In a recent op-ed, Norman Ornstein throws cold water on speculation about an independent Michael Bloomberg candidacy, as well as on the more general notion that there is an electoral market unserved by the existing two-party duopoly. Many Americans share Ornstein’s dismay about the condition of our politics; however, his jeremiad against the consequences of an independent presidential campaign goes beyond what the data can support. Public opinion data suggests that a run by Bloomberg – and/or Webb –would appeal to significant segments of the American electorate and thereby reveal the restricted representation provided by the existing two-party duopoly.
As Ornstein and his colleague Tom Mann clearly recognize, a decades-long process of party sorting has given us two ideologically distinct parties (what they call “parliamentary style” parties). Voters have clearer choices than in the past; there is little evidence that voters are happier with those choices, however. Many observers discount the expanding pool of self-identified independents, claiming that they are only partisans in independents’ clothing. These claims typically rest on a few studies—by now, decades old, and overlook substantial contrary evidence. Moreover, as Ornstein notes, recent research finds that partisans don’t like their own party any better than they used to, they just like the other party less.
Public opinion research has repeatedly found that on many issues the electorate is between the two parties. On abortion, for example, 80 percent or more of the electorate would prefer something between the “never, no exceptions” platform of the Republicans, and the “anytime, for any reason” position of the Democrats.
More generally we can think of political issues as being clustered into three categories – economic and social welfare (including taxes, social security, health care, income inequality etc.), cultural (including abortion, gay marriage, gun control, etc.) and foreign and defense policy (the most heterogeneous, including Iran, Syria, Ukraine, etc.). For simplicity assume that voters can take either of two sides on each cluster: an aggressive (A) or cautious (C) stance on foreign policy and defense issues, a left wing (L) or right wing (R) stance on economic issues, and a progressive (P) or traditional (T) stance on cultural issues. In this simplified universe, there are eight possible issue profiles (2X2X2) a citizen could hold:
Today’s Democratic Party offers voters the first profile, and today’s Republican Party the eighth. When the parties were more heterogeneous, local and state candidates would differentiate themselves from the national party when local or state issue profiles conflicted with the national profile (which was less well defined than it is now). But with today’s nationalized party system New Jersey and North Carolina Democrats look the same on the issues, as do Ohio and Oklahoma Republicans. This means that voters who hold any profile other than 1 and 8 (and research by Carmines and his collaborators indicates that many Americans do hold such profiles), must choose between two parties, each of which they disagree with on one or more issue clusters. But a Webb candidacy would appeal to voters with profile 2 and a Bloomberg candidacy to voters with profile 7. Voters in multi-party systems typically enjoy such a richer array of choices.
Independent candidacies would reveal how competition between two well-sorted parties unreasonably restricts the choices of the American electorate. As Ornstein notes, there is a downside to allowing voters a wider array of choices. If a president were elected with 35 percent or even less of the popular vote, perhaps after messy congressional bargaining, would he or she lack legitimacy and perhaps a congressional majority? Sure, but since the 1960s presidents often lacked congressional majorities. And such an outcome might teach the true believers in the Democratic and Republican parties a valuable lesson: that when they nominate a “real” liberal or conservative, there is no hidden majority that will turn out and sweep them to victory. As for electoral chaos, it would be temporary—the institutional features of our constitutional system make a two-party equilibrium the general condition of American politics. The likely outcome of serious independent candidacies and the resulting disruption would be to encourage a new generation of politicos to think about how to construct new electoral coalitions that would better fit the views of the American public and hopefully form a new American majority.
Morris P. Fiorina is the Wendt Family Professor of Political Science at Stanford University and a Senior Fellow of the Hoover Institution. He is currently writing a new book, “Parties at War: Partisan Sorting and the Contemporary American Electorate ” with Samuel J. Abrams.