Apropos the Swedish General Election

Tomorrow the Swedish General Election takes place. It got me wondering. In order to get elected as a Prime Minister in Sweden, you need to be the leader of a party. In order to be the leader of a party, you need to have risen through the ranks of that party.^[1] In order to rise through the ranks of a party, you need to spend time in that party. So given that we Swedes don’t usually elect seniors^[2], joining a party at a young age is a precondition for getting elected as Prime Minister.

So I see:

• Nooshi Dadgostar (Left Party) joined the Youth Left at age 14.
• Jimmie Åkesson (Sweden Democrats) decided to become active in the Sweden Democratic Youth Party at age 15.
• Ulf Kristersson (Moderate Party) formed a Moderate School Association with friends at his upper-secondary school, making him 15-19 at the time.
• Magdalena Andersson (Social Democrats) joined the Social Democratic Youth Party at age 16.
• Annie Lööf (Centre Party) joined her party at age 18.
• Ebba Busch (Christian Democrats) took employment at her party in 2006, making her 18-19.
• Johan Pehrson (Liberals) was vice president of Liberal Youth starting in 1991, meaning he probably joined at <21.
• Per Bolund (Green Party) entered politics at age ~35, though I don’t know when he decided that the Green Party was his.
• Märta Stenevi (Green Party) entered politics at age ~38, though I don’t know when she decided that the Green Party was hers.

I can’t even imagine what it’s like to pick a party as a teenager and stick to it for decades! I know I’m an outlier, having voted for five different parties across five national and three European elections. But I was clueless as a teenager and in my early twenties. Even when I held a good opinion, I did it for the wrong reasons. (I think I’m less clueless now, but of course I would.) I bet these nine – or seven if we exclude Bolund and Stenevi – were and did too. You just don’t know enough, and are not experienced enough, to be a good judge of political matters at that age.

You might think it’s rational to spend your whole career with a single party. While most 18-year-olds don’t know much about policy or macroeconomics, they do know their core values. Do core values change over time? If you look at annual polls on questions that rely heavily on value judgments (the permissibility of divorce, for example), it looks as if people change their values, but this could also be due to generational shifts. If people do change their values, it’s possible that they do so due to changing material conditions, in which case the party’s values may move in tandem.

I think there’s something to this, but if this were the whole story, that would mean voters were just as loyal to their party as politicians are. That doesn’t seem to be the case.^[3] So maybe there is something special about politicians after all.

One possibility is that you need to be a close-minded soldier in order to succeed as a politician (and close-mindedness and soldier mindset would drive a person to remain in one party). By “close-minded soldier” I mean something like “if you sat the politician down with an average person of their same intelligence and education level and gave each of them a battery of forecasting questions about a range of subjects, the politician would do markedly worse”.^[4] But if we measure this by the Big Five factors openness to experience (of which curiosity and adventurousness are parts) and disagreeableness (of which defiance and immodesty are parts) it actually seems like politicians are less close-minded than ordinary citizens. I found three relevant^[5] studies:

• Caprara et al. (2010) looks at personality differences between Italian MPs and Italian voters (conditional on age and education level) and finds that politicians are more open to experience (adjusted mean 3.78 versus 3.46 on a 1-5 scale) and more agreeable (adjusted mean 3.65 versus 3.40).
• Best (2011) (working with a sample one order of magnitude larger) looks at differences between German MPs and ordinary Germans and finds that politicians are more open to experience (0.89 higher mean on a 1-5 scale) and less agreeable (0.75 lower mean).
• Dynes, Hassell, and Miles (2018) looks at differences between public officials and ordinary citizens in the US and finds that “individuals with higher levels of […] openness are more likely to consider running for office, while agreeable […] individuals are significantly less interested”.

Cognitive biases could also play a role. Politicians, like everyone, engage in motivated reasoning (Baekgaard et al. 2017), and are as susceptible to sunk cost and status quo bias as the rest of us (Sheffer et al. 2017). But it’s not clear to me that politicians are worse than ordinary citizens in this regard.

Now it’s possible that politicians are likelier than ordinary citizens to lie on surveys so as to put themselves in a better light. Even granting that, I think the evidence for politicians being unusually soldiery looks weak. So I’ll tentatively discard the “has soldier mindset → becomes politician → stays in party” and “becomes politician → adopts soldier mindset → stays in party” models.

Another model: there’s a strong pressure to conform to the party line, locking politicians into the first party they join. So if you survey politicians and ordinary citizens on political matters first in a forum where they might expect others to see their responses, and then fully anonymously, you’d see a greater difference between public and anonymous answers among politicians than among ordinary citizens. That’s one way to think about it, but politicians could also internalise the party line, in which case the party acts as a regulator on its members’ sincerely held views.

I don’t know about this one. You’d expect a politician, especially a leader or rising star, to be under a lot of pressure to hold on to the party’s (and the party’s voters’) views. If I imagine a Swedish politician suddenly transported into ~Sweden, a country equivalent to Sweden in every respect except that the politician is unknown there, I expect the politician, now free from any external pressure to conform, would align themself with a different party some of the time at least.^[6] The causal model here is “becomes politician → feels pressure to conform → stays in party”. But I’m unsure about this.

What I’d like to know is if there’s something here that causes political leaders to make worse decisions than they otherwise would have – either a direct driver of bad decisions (e.g. pressure to conform) or a factor that causes us to select politicians that are likelier to make bad decisions (e.g. selecting close-minded soldiers). What I’ve discussed here are drivers related to political leaders’ marrying themselves to their party unusually early, but there are many instances of both kinds of drivers, and both of factors that drive good and bad decisions.

References #

Footnotes #

1. Alternatively, you can found a new party, like Gudrun Schyman did with Feminist Initiative. But I’m not aware of a case where a newly founded party has gotten anything close to a plurality vote in any Swedish election. I think this sort of thing is more common in Italian politics, which makes me wonder why the arrangement of parties is more plastic there. ↩︎

2. The oldest leader of a party with seats in the Riksdag is Ulf Kristersson at 58; after him comes Magdalena Andersson at 55 and Johan Pehrson at 54. The youngest is Ebba Busch who’s 35; she’s followed by Nooshi Dadgostar at 37 and Annie Lööf at 39. ↩︎

3. This phenomenon, however, could also be caused by generational shifts. If a party’s members have different demographics than the party’s voters, the party’s values can change as an older generation of politicians retires, forcing some of the party’s voters, whose values are no longer well represented among the party’s leadership, to support another party that’s better aligned with their values. ↩︎

4. This is of course not a perfect definition. I do think having true beliefs, being able to reason probabilistically and change one’s mind if the evidence warrants it are both components of a scout mindset and useful skills in forecasting, so I expect results on forecasting questions to correlate with scoutiness. Another way of testing for scout mindset might be “degree to which one achieves one’s goals”, but successful politicians are by definition successful – that’s how we selected them – so that seems like a bad way to figure out if politicians are unusually soldiery.

   Another question is whether a politician who privately holds perfectly true beliefs, but publicly holds false ones, is a scout. I think so. They are just also intellectually dishonest, but this seems orthogonal to scoutiness. ↩︎

5. These seem like exactly the sorts of studies that would produce different results in different parts of the world. I have no idea whether any of these observations (if they are right at all) generalise to non-Western populations. ↩︎

6. When imagining this thought experiment, I had to stop myself from stipulating that the parties of ~Sweden also had different names and logos than their counterparts in Sweden. I think that may actually make a difference. If so, that’s presumably because politicians come to identify with their chosen party, which becomes dear to them, and this is another factor that causes them to stick to a single party.

   Here’s another thought experiment. Say we replay the lives of these leaders, but at the point where they would’ve joined their party, they instead through some fluke join another party. Maybe they do it to impress a boy or girl that they like. So Jimmie Åkesson, for example, joined the Christian Democrats at 15 so that he could date some pretty Pentecostalist. Would he revert to the Sweden Democrats? How contingent is all this? ↩︎