A brief essay on polarization and constitutional structure
Every election season, experts sound off on the effects of the increasing polarization of our political parties. By virtually every standard, polarization has increased dramatically since the 1970s – whether you look at Congressional voting patterns, elite popular opinion, or language used by partisans – the parties are moving ever further apart. This year’s leading entry in this field is Thomas Mann and Norman Ornstein’s new book, It’s Even Worse Than It Looks: How the American Constitutional System Collided With the New Politics of Extremism. It argues that the American constitutional system is a poor fit for our newly polarized parties, particularly for what they describe as the one-sided polarization created by the radicalization of the Republican Party. A system with two legislative houses, a filibuster and a weird appointment process just can’t work when the parties, and particularly when one party, would rather bicker and blame than work together to govern. Legal scholars have contributed a great deal to this literature as well: Sandy Levinson’s great new book Framed and Rick Pildes’s fantastic Jorde Lectures from 2008 address the conflict created by the exigencies of modern politics and modern parties and our constitutional structure.
I think this literature is extremely useful for understanding our politics. Mann and Ornstein provide an important counterpoint to arguments by Responsible Party Government scholars of the 1950s who argued that clearly defined parties would provide voters with clearer choices, allowing popular governance even where voters have limited knowledge about many issues and politicians, and that strong, ideologically coherent parties are the only tool that can allow majorities to overcome the difficulty of getting legislation through our multi-step legislative process.
You can read these arguments and come to your own conclusions. But what I want to do here is bring some comparative perspective to this disagreement. My somewhat-fleshed out but still quite provisional argument proceeds in two parts: (1) Polarization in the United States can be understood as the way our political system integrates changes in popular opinion that are similar to what we have seen in other countries and that have resulted in radical parties in proportional representation (PR) systems and minority governance in Westminster systems; (2) These changes have been difficult for other constitutional/institutional systems to deal with as well. I’ve used blog posts to do some of this work before – see this and this – but current politics provide more ballast and flesh for this argument.
One way to think about polarization in comparative perspective is to imagine that public opinion that is unidimensional and peaked in the center – like a bell curve around a moderate opinion – and then imagine some shock occurs. Opinions get weird for one reason or another and cease taking an ordinary form. For our purposes here, it doesn’t matter whether it is symmetrical (with peaks on the right and left edges), asymmetrical (as Mann and Ornstein argue has happened in the US, with movement towards the right but not left edge) or just ceases to be unidimensional (with people starting to care about issues that do not correlate with the main left-right dimension in politics.) Just imagine opinion gets weird in some way, or just less centrist and unidimensional. As it gets weird, consider what might happen under three different systems: PR systems, Westminster systems like Britain, and the American system, with districted elections, a President and, crucially for my purposes, relatively easy to enter primaries.
The easiest one to see is PR. If opinion gets weird, parties will get weird too – as long as it’s enough to exceed thresholds and what have you, parties will form to attract newly radical opinion. If opinion gets weird enough, what you’ll see is the rise of really odd parties, that are either focused on strange issues or that are sufficiently radical that mainstream parties can’t form coalitions with him. Have we seen this? Yes, quite frequently. From far-right anti-immigration parties like the Sweden Democrats to left-wing groups like Die Linke in Germany to really scary things like Greece’s Golden Dawn to just plain old weird parties like the international Pirate Party, we have seen the rise of new political parties that win seats but exist outside of even the capacious definitions of the mainstream ordinary politics in Europe.
Westminster systems like Britain and its former colonies use single-member districts and first-past-the-post (plurality wins) vote counting. This, following something called Duverger’s Law, is supposed to lead to a stable two party system. Because neither voters nor political figures like donors or candidates like wasting their effort, they focus on potential winners. However, if you insert a few assumptions – like people get some expressive value from voting – this can change in the face of weirdness in popular opinion. If opinion gets sufficiently on the fringe, the difference between the main parties can seem less relevant and politicians and voters may turn to third party efforts. Have we seen this? Yes, both at the district and national level. There was a period when there were no majority governments in any Westminster system — that is, no parliament where a party held a majority of the seats. It’s not true anymore, as Steven Harper’s Conservatives won a majority in Canada in 2011. Yet none of the major Westminster countries — including Canada, which is either a three or four party system depending on which province one is in– are real two party systems. In most Westminster countries today, there are three or more parties seriously contesting each district.
Finally, think of the U.S. We have the Duvergerian impulses created by single-member districts and first past the post vote counting, and on top of it, we have a powerful Presidency, another force encouraging a two-party system. We also, however, have quite open primaries, allowing groups to shape party coalitions more easily. What would a shock to popular opinion do in such a system? One might imagine that whatever radical groups emerged would make efforts to take over the national parties through involvement in primaries, by running and supporting candidates with an eye on something other than choosing the most electable candidate, particularly in non-Presidential elections (where Duverger’s Law’s weight might be felt more fully). This would be a primary-level version of the third-party impulse we see in Westminster systems. Sound familiar? Tea Party candidates, from Christine O’Donnell and Sharron Angle last cycle to Richard Mourdock this one, have made it harder for Republicans to win seats. But electability wasn’t the major concern of their supporters in primaries.
This is hardly determinative of, well, anything. But it is suggestive of the possibility that something common (or at least of common form) has happened across the developed world. (Why opinion would have radicalized everywhere is an interesting question but I can’t do much more than engaging in armchair speculation on it.) However, our analyses of the causes of polarization are not complete without considered the possibility that it is not a purely American phenomenon, but rather is the way American institutions integrate a change that has happened to different degrees and in different ways across the western world.
There is a long literature on the comparative benefits of Presidential and parliamentary systems – Juan Linz’s pathbreaking work stands out particularly – and I don’t mean to enter the long debate on the best forms of government. My point here is just to note that the effects of increased variance in popular opinion have created constitutional problems in many countries, not just ours.
One of the basic ideas of a PR/Parliamentary system is that elected officials are supposed to form majority coalitions. Voters don’t choose among coalitions that can govern by themselves – they vote for narrower parties who then negotiate among themselves to form a government. If the effect of increasingly diffuse opinion to create parties that are too radical for mainstream parties to join in coalitions with, this makes it increasingly hard to form governments. Some examples: Belgium’s failure to form a government for more than a year; the difficulty the German Social Democrats and Greens would have forming a government due to the influence of the Die Linke; numerous grand coalition governments; France’s 2002 election where the National Front made the Presidential runoff. All of these can be thought of as constitutional problems akin to the difficult polarized parties in the U.S. have at solving debt problems when the very different parties divide Congress and the Presidency.
Westminister systems are designed to create referenda on governing parties – as I’ve noted elsewhere, Duverger’s Law is normative as well as positive. But when you have three or more parties, you don’t necessarily get majority results. The last Canadian election was won with 37% of the vote. This doesn’t create governing problems; but it could create substantial legitimacy problems if a party won a majority without being more popular the combined support of two ideologically similar opponents.
I’m not sure whether our problems are worse or better than the problems in these other countries. And I certainly don’t think we should assume our Constitution is the best of all possible Constitutions merely because it is ours and has lasted for a long time. But I do want to stress that governing in a world where popular opinion is very divided is extremely difficult, no matter what constitutional or institutional system exists in a country.