Majoritarianism with little strategic voting in Andorra

The Pyrennean microstate of Andorra held a parliamentary election in March.

The principality’s General Council consists of 28 members elected by a majoritarian mixed-member system (MMM): half is elected proportionally from national party lists (largest remainders; the threshold is one Hare quota – 7.15%) and the other half, in parallel, by list plurality (party with most votes takes all seats) – 2 from each of the country’s seven parishes, with quite a bit of malapportionment.

The results were as follows:

Party District List PR Total
Votes (%) Seats Votes (%) Seats
Democrats for Andorra 39.4 (38.6) 10 37.0  (36.3) 5 15
Liberal Party of Andorra 27.5 (26.1) 4 27.7  (26.7) 4 8
PS+VERDS+IC+I 23.6 (24.0) 0 23.5  (24.6) 3 3
Social Democracy and Progress 9.5   (11.3) 0 11.7  (12.5) 2 2
  • The figures in brackets exclude votes from the two parishes where one or more parties did not put forward a slate of candidates.
  • The Democrats and Liberals were allied with various local parties and independents in most parishes.

Curiously, though the tiers are clearly separate, the parties’ vote shares in the different tiers were almost identical (a total of 4.9% when adding up the differences, 4.7% when disregarding votes from districts with incomplete nominations). The same pattern largely holds across districts, and can be observed in previous election results. The only times there has been a significant difference between vote shares in the two tiers occured when certain parties did not nominate in a district.In the districts, the Democrats and Liberals were allied with different parties and independents.

These results run counter to the expectation of strategic voting in the district tier, where the smaller parties had little chance of winning seats. Such strategic voting (in one tier) is common in mixed-member systems. If we can call the pattern in Andorra’s results cross-tier contamination, it would seem to run from the list tier to the district tier rather than the other way. However, this may not the best way to look at it.

Maybe the lack of polls in this small country left voters with too little information to vote strategically? But they could still have learnt about the different parties’ chances from previous elections (of course, strategic voting elsewhere predates the proliferation of public opinion polls anyway). Andorran parties certainly seem to have nominated more strategically in the past, most notably in the previous election (2011) when only two parties competed in each district; this was more rare this time, but alliances between parties remain common.

It may not be of any great significance, but I think it’s a fascinating anomaly.

5 thoughts on “Majoritarianism with little strategic voting in Andorra

  1. Thanks, JD. This is interesting.

    One could quibble with whether the Andorran system should be called MMM. Shugart and Wattenberg’s definition, at least as I recall it (it has been a while!), would say no. The definition there for a “mixed-member” system is that there must be a nominal tier and a list tier, without regard for for whether the allocation formulas in the respective tiers are majoritarian or proportional, although it is MMM only if the nominal tier uses majoritarian rules of some sort (and MMP only if the list tier is compensatory).

    If the “district tier” here is, as JD says, a list, then there is no nominal voting. Now, with M=2, maybe that is a distinction without a difference, and the S&W definition is overly narrow. But if we imagined the same system but with M=5 (for example), then it would be hard to say it is effectively nominal. Where to draw the line? Well, S&W draw it not on magnitude, but based on how the members are elected: via their own votes cast exclusively for them by name (hence via “nominal” rules) or tickets presented by groups across whose candidates votes are pooled (hence via “list”).

    The precise categorization may not be important, except that perhaps the use of lists in the districts could help explain the surprising lack of strategic voting. I wonder what the ballot looks like. If both parts just show party symbols, it might be “psychologically” less likely to induce strategic behavior than if voters are made aware that they are selecting candidates in (small) districts (even if party is the main criterion of such selection).

    I do not know, just speculation, or what in fancier Greek terms might be called a hypothesis. One I can’t test, alas!

    Thanks for calling this unusual system to my attention. This is exactly the sort of thing for which I keep this blog running.

  2. It is, however, noteworthy that the number of competitors in the districts increased. So do I understand correctly in the past that there would have been more choices on the national list ballot than for the district lists? Obviously that is strategic behavior by elites. So I guess the bigger question than the one about strategic voting might be why was elite behavior different this time.

    I will admit I have had a longtime fascination with Andorra, although until today not with its electoral system. But the concepts of a co-principality and a state up in the mountains always fascinated me. I hope to visit some day.

    • The previous election (2011) saw no more than 2 lists per district, but in 2009 and 2005 most districts had three or more lists, though the pattern of nominations wasn’t quite as close to uniform as this election.

  3. Perhaps one explanation for this phenomenon could be that while the Democrats are centrist, and the Liberals are right-wing, the other two parties, the voters of whom would be expected to act tactically, are left-wing. I don’t claim to be an expert on Andorran politics, but maybe the voters for these two parties hate the Democrats and Liberals so much that they are prepared to ‘waste their votes’, so to speak, rather than vote for one of them.

    • Strange that they don’t vote tactically among the two parties of the left. At least one district where they would seem to have had a realistic chance of electing one of them, had they voted en bloc.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.