Responsible Parties: Saving Democracy From Itself
Compounding the Founders’ Folly
It is no surprise that the American system has weak parties. It was designed by people who abjured the very idea of parties. George Washington denounced them as incubators of factionalism, a view shared by James Madison and Thomas Jefferson, who declared in 1789, “If I could not go to heaven but when a party, I would not go there at all.”60 But the Founders soon changed their minds. In the early 1790s, when it became obvious that they could not stop Alexander Hamilton’s centralizing agenda without an organized party, they formed the Democratic-Republicans to defeat the Federalists – which they did in 1800. But they were still operating under a Constitution whose framers had embraced the conventional distrust of the very idea of organized political dissension that they had imbibed in the Old World.
The Whigs and Tories had set up opposing camps in Parliament a century earlier, but at the time of the American Revolution eighteenth-century Britain was in a long phase of Whig-dominated cabinet government that the Tories were powerless to resist. Henry St. John, better known to us as Viscount Bolingbroke, who led the Tories during their years of disgrace following a failed attempt to bring back the Stuart dynasty in 1715, extolled the virtues of nonpartisan government. It made sense for Bolingbroke to oppose parties when his party was down. Richard Hofstadter notes that the American revolutionaries had been raised “on imported criticisms of the corruption of Walpole’s era.”61 It was natural for them to bring this distrust of partisanship with them across the Atlantic and embed it in their Constitution.
But the Founders' resistance to strong parties had deeper roots than Bolingbroke’s belief, mocked as naïve by Edmund Burke, that it is possible to have politics without partisanship. Their aspirations were fundamentally at odds with Burke’s approving view of the political party as “a body of men united, for promoting by their joint endeavors the national interest, upon some particular principle in which they are all agreed.”62 They held an abiding hostility to the idea that the federal government should be able to promote anything beyond the most rudimentary conception of a national interest.
The American republic was a defensive response to the costs of the revolution. The war against Britain had dragged on for years and almost been lost several times due to the government’s inability to get recalcitrant states to supply funds and troops when needed. After the war, the government faced the same impotence in raising funds to service – never mind pay off its war debts. This stymied basic government operations and made American participation in international trade virtually impossible. The Founders wanted to solve these problems by creating a national government with the resources to defend the nation and finance itself, but they were determined to stop it from pursuing more extravagant conceptions of national purpose. This is why they insisted on the Tenth Amendment, which reserves all powers not expressly delegated to the federal government to the states, and why they were so obsessed with limiting congressional power. The federal government was not meant to do much.
So we inherited a system designed by people with little interest in a government that would pursue national political programs and even less interest in competing political parties. Once parties did emerge, they quickly became centers of gravity of political competition, and though the Founders knew nothing of the dynamics of electoral politics, their choices gave us two. This is because of Duverger’s Law, which holds that the number of parties will be one more than the number of representatives elected in each district in a given election. Other factors, notably constituency size, primaries, runoffs, and regional variation in political attitudes, can counteract Duverger’s logic, but in the U.S., while such factors have sometimes put great strains on the two-party system, they have never destroyed it. If the number of parties was all that mattered, the U.S. would have a healthy system with two programmatic parties.
But party strength matters as much as party size, and here the founders came up short. The independently elected president, the staggered electoral cycle, and bicameralism all weaken the congressional parties and their leaders. Representation in the Senate is especially problematic. In 1790 Virginia’s population was 12.65 times that of Delaware. By 2010 California’s population was 76 times that of Wyoming.63 Today, seven states have such small populations that they qualify for only a single House member, yet Article I of the Constitution gives each of them two senators.64 This means that while one-quarter of the country’s population is represented by sixty-two senators, only six senators represent another quarter – an imbalance that will only worsen as growing urban populations accentuate the built-in advantage of small rural states.65
Many aspects of today’s American system are unintended consquences of the Founders’ intentions, but with the multiple veto they got exactly what they wanted. The legislature’s internal design, which makes it difficult to act, is compounded by external constraints: the other branches in a separation-of-powers system and strong federalism. The large number of veto players creates a powerful bias toward the status quo, and toward those who have enough resources or who are strategically well placed to manipulate the status quo. The net result is to undermine legislators’ accountability to voters, feeding their alienation and fueling demands for more direct democracy.
But the Founders' unfortunate choices have been compounded by subsequent generations’ misguided if well-intentioned reforms. Among the prime culprits were the Progressives. Early in the last century, their commitment to bringing democracy closer to the people gave us primaries and caucuses that weakened party leaders and worked against creating coherent national party platforms.66 Parties did strengthen leaderships at the expense of committee barons to some extent in the middle part of the century. But the centralizing reforms proved less than equal to the task, and after 1968, the McGovern-Fraser Commission ushered in a new era of enhanced grassroots control marked by the greater importance of primaries and caucuses and the advent of descriptive representation in the selection of party delegates. The architects of those reforms were right to wrest control from party elites and hacks, but instead of empowering congressional backbenchers, they turned power over to activists, who participate disproportionately in primaries and caucuses, and to organized groups that can fund and capture them. The net effect has been to weaken further already weak parties.
Decentralization of presidential selection has compounded these problems. Both parties have some delegates at their national conventions who are party officials, and whose presence is meant to counteract the centrifugal effects of primaries and caucuses. The Republicans have three per state, and although they are technically uncommitted, since 1980 there has been a strong presumption that they will deviate from their state’s primary or caucus result only if the first round of convention voting is inconclusive.67 This presumption was tested in 2016, and it proved sufficiently strong that they could not stop Donald Trump’s hostile takeover of their party despite various schemes that were mooted to do exactly that until the eve of the convention.
The Democratic establishment did manage to stop Bernie Sanders. The great majority of their superdelegates – party officials and activists who constitute some 15 percent of the total – pledged their support to Hillary Clinton well before the primaries ended. This did not decide the outcome, in that she ended up with enough support without them, but their early pledges helped create momentum and funnel money her way – which many Sanders supporters believed gave her an unfair advantage. The cost of Sanders’s eventual support for the Clinton-Kane ticket was an agreement to future reforms that will bind most superdelegates to the primary and caucus results.68 This will further untether presidents from their parties, making it harder for them to operate as surrogate prime ministers. It used to be conventional wisdom among political scientists that the importance of primaries since the 1960s has been overrated because party establishments preselected candidates in an “invisible primary” before the voting began.69 This was always debatable, as Stephen Gardbaum and Richard Pildes have noted, and the recent nominations of John McCain, Barack Obama, and Donald Trump (not to mention the price that the Democrats have paid for wresting the 2016 nomination from Bernie Sanders) make it clear that the party no longer decides – if it ever did.70
The dissatisfaction with superdelegates is one manifestation of antipathy for the undemocratic way in which presidential candidates are selected. Another is the Electoral College system, whereby the great majority of states cast all their votes for the candidate who wins the popular vote in their state. This can produce the kind of anomaly that occurred in 2016, when Trump won the presidency with 304 Electoral College votes to Clinton’s 227, even though she won the popular vote by more than 2.8 million votes or 2.1 percent.71 This prompted calls for states to adopt a “national popular vote” blueprint that would bind them ex ante to cast all their Electoral College votes to whichever candidate wins a plurality of the national popular vote.72
Winning the presidency while losing the popular vote undermines a president’s legitimacy, as Trump underlined by insisting doggedly that he actually won the popular vote when what he alleged to be millions of illegal Democratic votes are discounted. His allegations prompted widespread derision, partly because there was no evidence of significant fraud and partly because the Trump campaign had strongly opposed recounts in several close states – insisting there was no evidence of fraud or other irregularity.73 The Presidential Advisory Commission on Election Integrity that Trump subsequently appointed in hopes of vindicating his claims soon collapsed into partisan warfare that wound up in the courts and was then disbanded.74 Trump’s almost comic desperation in this matter was fodder for late-night television, but the more important issue this episode underscores is that strengthening the president’s legitimacy by reforming the Electoral College system is a bad idea. It would be better to weaken it further and strengthen congressional leaders instead.
60. Thomas Jefferson to Francis Hopkinson, March 13, 1789, Founders Online (accessed 12-18-2017).
61. Richard Hofstadter, The Idea of a Party System; The Rise of Legitimate Opposition in the United States, 1780-1840 (Berkeley: University of California Press, 1969), 19.
62. Edmund Burke, “Thoughts on the Cause of the Present Discontents,” in Select Works, vol 1 (Indianapolis: Liberty Fund, 1999 ), 150.
63. Dylan Matthews, “You Can’t Understand What’s Happened to the Senate Without These Two Graphs,” Washington Post, April 8, 2013 (accessed 11-10-2017).
64. They are Alaska, Delaware, Montana, North Dakota, South Dakota, Wyoming, and Vermont. U.S. Government Information (accessed 11-10-2017).
65. The Small-State Advantage in the U.S. Senate, New York Times, March 10, 2013 (accessed 11-8-2017).
66. Rosenblum, Nancy M. On the Side of Angels. Princeton University Press, 2010, 165-253.
67. See Kerry Picket, “RNC Rules Committee Member: Every Delegate at GOP Convention Not Bound on First Ballot,” Daily Caller, March 13, 2013 (accessed 11-11-2017).
68. David Weigel, “Democrats Vote to Bind Most Superdelegates to State Primary Results,” Washington Post, July 23, 2016, (accessed 11-10-2017).
69. Marty Cohen et al., The Party Decides: Presidential Nominations Before and After Reform (Chicago: University of Chicago Pres, 2008).
70. Stephen Gardbaum and Richard Pildes, “Populism and Democratic Institutional Design: Methods of Selecting Candidates for Chief Executive in the United States and other Democracies,” NYU Law Review (2018): 17-22.
71. Federal Election Commission, “Official 2016 Presidential Results,” (accessed January 30, 2017).
72. This is not to be confused with the congressional district method used in Maine and Nebraska, where the state allocates one electoral vote to each district, awarding it on the basis of that district’s electoral vote. The remaining two electoral votes are awarded to the winner of the statewide electoral vote. Fairvote Archives: Maine and Nebraska (accessed 11-11-2017). If adopted nationwide, this system would likely work to the disadvantage of Democrats, given the concentration of their voters in urban districts.
73. Monica Davey, Julia Bosman, and Steve Eder, “Trump Backers Go to Court to Block Vote Recounts in Three States,” New York Times, December 2, 2016 (accessed 11-11-2017); Michael Tackett and Michael Wines, “Trump Disbands Commission on Voter Fraud,” New York Times, January 3, 2018 (accessed 2-12-2018).
Frances McCall Rosenbluth is the Damon Wells Professor of Political Science and Ian Shapiro is Sterling Professor of Political Science, both at Yale University. Shapiro is also the Henry R. Luce director of the MacMillan Center at Yale Univeristy and the publisher of YaleGlobal Online.