Would We Be Better Off With Non-Ideological Centrist Parties?

by Michael Dorf

In my new Verdict column, I discuss the phenomenon of political "pivoting," whereby a Republican runs to the right in the primaries and to the center in the general election, whereas a Democrat runs to the left in the primaries and to the center in the general. I argue: (1) pivoting is a kind of dishonesty that is made virtually inevitable by party polarization; (2) our very long presidential election campaigns facilitate pivoting because there's enough time between the primaries and the general for low-information swing voters to forget what happened in the primaries; and (3) because the media treat pivoting as normal, it falls to opposing candidates to point out inconsistencies, which, in turn, requires money for advertising, and thus increases the influence of money on elections.

Here I want to briefly explore a related issue I mention in the column: the greater intra-party coherence and inter-party polarization that has developed over the last generation or two. Political scientists debate the causes and the magnitude of the effect, but no one seriously doubts that it is real. Even Morris Fiorina and his co-authors, who have argued (in this book and also this one) that the American people are far less polarized than our politicians are, do not deny that the parties are more polarized than in earlier eras. For present purposes, I'm going to take as given that across a range of issues, the distance between the center of the Republican party and the center of the Democratic party is greater than the same distance measured a generation ago. The question I want to ask is whether this is good or bad.

Fiorina et al think that polarized parties against a backdrop of a largely moderate electorate is almost self-evidently bad. They contend that as a result of the polarization of the political class, politics has become a contest over largely symbolic issues that matter to a small number of activists, while the issues that matter most to the general public are largely ignored. I don't think that this is an entirely fair characterization of our politics, but even if it is, it's not an inevitable consequence of polarization.

Quite the contrary, one would think that polarized parties would have more of important substance to disagree about than would two centrist parties. If one party is pro-environmental regulation, pro-progressive taxation, and for civil rights and civil liberties, whereas the other is laissez-faire, for low taxation, and for law-and-order policies, then elections should be fought over high-stakes questions. It's precisely when there are no clear ideological differences between the parties and their platforms that politics is likely to become dominated by questions about personality and exaggerated differences on symbolic questions.

Does it follow that we should welcome polarization as at least a potential antidote to the personalization and trivialization of our politics? Not necessarily. Centrist parties have two clear advantages. First, they tend to lower the stakes of politics, which will often permit the formation of policy based on the best evidence (where the underlying issues have empirical components). Second, centrist parties ensure that policy is always close to the median of public opinion. By contrast, with polarization, policy can swing from one extreme to the other.

That is the central complaint of Fiorina et al: that the silent middle is left largely unserved. To the extent that the complaint is fair, it points to a serious defect. Representative democracy does not require that policy always track the center of public opinion. Certainly there is room for elected officials to chart a Burkean course by voting for what they regard as the best policy, on the assumption that the voters choose them to exercise independent judgment, not simply to act as conduits of public opinion.

However, as a practical matter, there are limits to any elected official's ability to depart too far from the public opinion of his or her constituents. Moreover, the whole premise of representative government is that the people should be heard and the majority should rule (subject to protections for minority rights). Working from first principles, one would not design a system of government to produce policy at one extreme or the other but never in the center of public opinion.

But, of course, that is not a fair account of American government. On the whole, our government in fact produces centrist policies. Partly that's because presidential candidates (and, eventually, presidents) have incentives to pivot, but even elected officials who never pivot (like Tea Party Republicans) don't yield wild swings from left to right because they don't act alone. In a parliamentary system with a dominant legislative chamber, a single election really can produce a radical shift in policy. But our system, with staggered Senate terms, power divided roughly equally between the House and Senate, and veto power in the president, means that it is extremely difficult to overcome the political inertia needed to enact divisive policies. Divided government undercuts the tendency towards the extremes of party polarization.

So, to answer the question that titles this post, we probably would not be better off with non-ideological centrist parties--at least insofar as the trouble with polarization is the production of personality-driven and trivializing campaigns and wild swings of policy from one extreme to the other. Our personality-driven and trivializing campaigns are a problem, but not one obviously caused by polarization, whereas we don't in fact see wild policy swings.


Postscript: From where I stand, we do not have a right-wing and a left-wing party. We have a very right-wing party and a center-left party. But I have written the foregoing on the assumption that what matters is the distance between the parties, not where they align on some absolute scale (or even some scale that compares the Overton window in America with its counterpart in other democracies).