Do Political Primaries Fuel Polarization and Extremism?

Rick Pildes

Yesterday's primary elections re-raise two questions about how the design of our electoral processes might further fuel the extreme partisan polarization that has increasingly characterized American democracy over the last 30 years. Primaries are frequently a cite where more extreme candidates are driving out more moderate ones (for better or worse -- this is indeed, a debatable question).

One question is whether primary elections themselves tend to do so; the second question is whether certain forms of primaries accentuate this tendency all the more. In Delaware, for example, the Republican Party uses a closed primary, in which only registered Republicans can vote. Nearly a quarter of registered voters in Delaware are independents (around 23.5%). None of them could participate, therefore, in Delaware's primary yesterday. The reality of our two-stage election process is that it's become hard enough to get voters to turn out for the general election. Turnout for primaries tends to be far below those levels. Moreover, those who tend to turnout, at least in some primaries, come not surprisingly from the most activist wings of the parties. Primary electorates with low turnout might thus not even be representative of the Republican (or Democratic) voters as a whole. In addition, if independents are legally shut out of the process, through closed-primary laws like in DE, and independents tend to be more moderate, centrist voters, then primaries will diminish the more centrist forces and empower the more extreme ones. Whether this is a good or a bad thing is debatable: for those of the view that sharply differentiated political parties makes for more informed voting, it might be a good; for those worried about the decline of the center, it would be a negative. There is also an empirical debate about how representative or not primary electorates turn out to be and whether closed primaries do in fact produce more extreme candidates than open ones.

But if one concludes the current structure of primaries is a problem for democracy today, either because turnout is so low and/or because primaries empower the extremes (and that hyperpolarized politics is not good for American democracy), there are at least two institutional design changes to consider. The most obvious is simply to push more states to use open rather than closed primaries. A more dramatic (and far less plausible, therefore, in practice) approach is to adopt systems of voting like instant-runoff voting. One feature of these systems is that they collapse the primary and general election into one day, so that the low turnout problem of primaries is avoided. I was recently asked to contribute to an interesting website called Big Think, for a series on "Dangerous Ideas," and I intentionally chose a provocative way of framing these issues by calling mine "Abolish Primary Elections." You can hear more about instant runoff voting, and issues about primary elections, there (the production quality of their video work, by the way, is exceptional). For an academic piece of work that presents these issues more fully and summarizes the existing empirical literature, see here.
Posted 9:26 AM by Rick Pildes [link]