In recent years, Dutch politics has fragmented. In part, this is a function of the country’s extreme form of proportional representation, which allows very small parties into Parliament. In addition, ballot access is easy. All you need to do is put up a refundable deposit of €11,250 and you’re on.
Nonetheless, the Dutch party system has seen increasing fragmentation in recent years. The standard measure of the effective number of parties takes each party’s share of seats, squares them, and adds up the result. A two-party system will always have a number below two, since one party or the other will have a majority. A system in which one party wins a majority will always have a number below four.
The number has issues, of course, all measures do. For example, it suggests that a parliament with three equally-sized parties (3.0 effective parties) is about the same as one with a majority party (53%) and five smaller ones (3.1 effective parties). Obviously whether that is true depends on the ability of those parties to form coalitions: the first system might be a competitive kaleidoscope while the second system could be a de facto one-party state. Or vice versa. In short, the measure does not tell you how easy it is to form governments or how different the possible coalitions on offer might be.
Still, the measure captures the degree of parliamentary fractionalization quite well. And you can use it to make comparisons across time. And so, here is the number of effective parties in the postwar Netherlands:
Rising fragmentation creates a problem for voters, of course: which party to vote for? There are more options than ever and many of those options are brand new.
Well, André Krouwel a political scientist at the Vrije Universiteit Amsterdam, developed an online survey to help voters with this problem. A quarter of Dutch voters used it in the 2017 election. Since this will be relevant for me in the next Dutch general election, I decided to give it a whirl.
And to my surprise, it pegged me as a Christian Union voter. Huh? I took it again, same result.
So I took a different survey. Now I was with the Greens (officially “GreenLeft”), which makes more sense.
What went wrong the first time out? Well, simple: the quiz did not properly weight the importance of each question. Consider the following two sets of political positions:
- Abolishing legal suicide
- Recriminalizing marijuana
- Toughening criminal justice
- Ending automatic organ donation unless you opt-out
and
- Opposing mandatory national service
- Abolishing health care deductibles
- Ending nuclear power
- Enabling referenda to overturn laws
- Liberalizing dismissal laws
The first group are policies with which I disagree with the Christian Union and agree with the Greens. The second group are policies about which which I disagree with the Greens and agree with the Christian Union.
With the exception of #4 on the first list, I feel rather strongly about these issues. With the exception #4 on the second list, I do not feel particularly strongly about these issues. So I am much closer to the Greens than to the Christian Union. But the election survey had no way to know this.
Aside: why am I unbothered about where I disagree with the Greens? Well, (1) is not going to happen so whatever. (2) is an issue, but the Greens recognize that premiums will have to rise; I disagree about the tradeoff but the Greens are not peddling nirvana. (3) is a major issue for Europe and the world ... but the Netherlands has only one small nuclear power plant. (5) is also a real issue, but the Dutch labor market performs all right regardless, so it is hard to get worked up either way.
There are a lot more policies on which the two parties agree, by the way. The Christian Union is a weird animal by American standards, being socially conservative but redistributionist, pro-immigration and very green. In fact, GreenLeft came into being from a merger of four parties, one of which was an openly Christian organization called the Evangelical People’s Party.
Anyway, to return to the title of this post, beware these political quizzes, including the ones that try to site you in some sort of two-axis issue space. They have trouble capturing issue valence and are therefore not to be trusted.
Comments