Podcast #124: Odds and evens in Spain and Australia

The 2023 Spanish general election was heavily tipped to be a victory for the centre-right People’s Party (Spanish: Partido Popular; PP). Instead it elected a hung parliament with the centre-left Spanish Socialist Worker’s Party (Spanish: Partido Socialista Obrero Español; PSOE) forming government with Catalan nationalists, despite PP winning the most seats and the most votes nationwide (because PP and the right-wing Vox party combined were six seats short of a majority).

A shocking victory despite all the baggage surrounding Pedro Sánchez (the Prime Minister and PSOE leader) and the fact that Albert Núñez Feijóo (the Opposition Leader and PP leader) was ahead in the polls. Potentially Pedro Sánchez had a personal vote that won him seats and votes but even then it was PP that had a majority in Madrid, where both leaders are from.

The left only won a few provinces (subregions of Spain’s autonomous regions), which includes all four provinces of Catalonia (Barcelona, Girona, Lleida and Tarragona). PSOE barely held onto Álava in the Basque Country and Navarre in Navarre, but they comfortably held onto Badajoz and Caceres in Extremadura and Seville in Andalusia. They also marginally held onto Las Palmas in the Canary Islands despite losing the neighbouring province of Santa Cruz de Tenerife.

NP, did you actually listen to the podcast? Got any thoughts on odd and even district magnitudes and uneven magnitudes?

Direct link to research, for those who dislike stumbling around – https://research-repository.griffith.edu.au/bitstreams/3daa1d3f-134d-4729-92fd-ae4d1e67111a/download

Excellent podcast once again Ben, thank you for creating and sharing original content with us.

It’s interesting to hear how electoral systems in other countries have developed. I imagine something like this would be near impossible to introduce in Australia due to our historical experience with single member electorates, and the desire to have a clear election winner in the Westminster system. However, if we were to suspend reality, I would hazard a guess and say that three member electorates would be favoured by the Coalition, while Labor, the Greens, One Nation and others would prefer a five member arrangement, especially if we used d’Hondt as in Spain.

Spain is clearly not a democracy, Sanchez should not be PM, The opposition PP won more seats, Nunez should have been sworn in as PM instead of Sanchez cheating and doing a dirty deal with the separatists in Catalonia.

He should immediately resign and advise the king to swear in the opposition or call a snap election. Does the map which is almost completely blue not tell you enough?

@Daniel T I agree that Núñez would make a better PM, and it is bad that a politician like Sánchez who like Núñez opposes Catalan independence has done a deal with Catalan nationalists.

@Ben Raue I didn’t listen much sorry I just wanted to provide some info about Spain for those who didn’t know. I like to share information as one who is knowledgeable.

@Daniel T and Neither Portal, this is a flaw with a European style proportional representation coalition and would fail in a hypothetical referendum in Australia as both major parties will oppose it (arguing with the Spain example) and Australians prefers a majority government.

@Marh it’s not the same in France where they use single-member constituencies with FPTP voting (with second rounds if and where necessary). In Germany they use MMP like New Zealand does, which is where single-member constituencies are used where the member is elected by winning the first ballot (known in NZ as the electorate vote) but sometimes one or more list members are elected from a constituency based on the second ballot (called the party vote in NZ) where other party candidates can get elected.

Italy, Poland and Spain all use multi-member constituencies with proportional voting. In Spain they are based on the provinces of Spain’s 27 regions which are equal to LGAs in Australian states and territories.

The D’Hondt proportional method (the system used in Fiji) is very popular in Europe, perhaps since it was invented there (well it was first described by an American, Founding Father and former President Thomas Jefferson, in 1792, but was reinvented by a Belgian mathematician named Victor D’Hondt in 1878).

Åland, Albania, Armenia, Austria, Austria, Belgium, Croatia, Estonia, Finland, Greenland, Iceland, Luxembourg, Moldova, Monaco, Montenegro, the Netherlands, Poland, Portugal, Romania, San Marino, Serbia, Slovenia, Spain, Switzerland and Turkey all use this method, while Hungary and Italy use a mixed system that is derived from the D’Hondt method. The D’Hondt method is also used for some seats in some British elections (top-up seats in the Scottish Parliament, the Senedd in Wales and the London Assembly) and by some countries in European Parliament elections, as well as by numerous countries outside of Europe.

At the 2023 Finnish parliamentary election, the right-wing Finns Party was two seats away from being the largest party in Parliament, and 0.7% away from winning the most votes at the election.

Daniel, it’s hard to take you seriously with such idiotic opinions.

Spain had a close election. Neither the left or right won a majority (on either the vote or the seats).

The electoral system actually boosted the right, who won 48% of the seats off 45% of the vote while the left won 43% of seats off 44% of the vote.

It’s a proper hung parliament, of the type you could see in a majoritarian system too. If that’s not a democracy, not many countries are.

If the right wanted to form government they should have won the election.

Also what on earth does the map show you, apart from the fact that rural areas vote for the right? Obviously the red areas have a much more dense population.

Sorry Ben, I was just trying to point out the largest party should always be the one to form government at least first. Because look at Canada for example. (They don’t use the term ”Hung Parliament” there) They always let the largest party form government, even on knife edge elections like 1972 in Canada. They let the government stay on despite the fact the other parties could have constitutionally voted them down.

Take 2006 for example in Canada, the left won a majority, Lib + NDP + BQ made a majority, even in 2008. Yet they didn’t form government, the conservatives did.

The same should apply everywhere, when you win the most votes and seats you should try to form a minority government and if they lose a confidence vote, then and only then let the 2nd biggest party give it a go.

There are very few examples in western democracies where the party trailing in seats forms government (outside of fringe extreme parties winning the most seats)

Also the only populated areas the Spanish socialists won was Barcelona/Catalonia. They lost Madrid. They did win some rural areas in western Spain however. But the map is Blue because the PP won more seats. as it shows a plurality.

The only other examples is Sweden and Norway since the Left are always the largest party but the center-right vote is always fractured. The 3rd place finisher in Sweden is the government oddly.

That is exactly what happened in Spain in 2023. The king let the the leader of the PP attempt to form a government first, and when that failed the PSOE got the chance.

You can let someone have a first go at forming a government but ultimately if you don’t have a majority you don’t have a right to govern – you need a majority to at least tolerate you.

Many democracies require a positive vote of investiture, not just being able to survive votes of no confidence.

Anyway this was straight forward, the current Spanish government won the support of a majority of members, despite their opponents being over-represented in the parliament.

The idea that the party with the most seats should necessarily be the one to form government is absolutely nonsensical. Naturally, they usually will be the one to form government, but they are not entitled to it. If applied to Australian politics, Labor should have formed government in 2016 and 2019!

The current Dutch government setting is probably the most bizarre with a four-party coalition (2 centre-right, 2 right-wing populist) but had to hire an independent PM (formerly from a centre-left party) to enable a cabinet

Quite interesting, though slightly ruined by repeated idea that right-wing parties do better in rural areas and the left in urban areas. This isn’t really true in Spain (it has become more true recently, but that is only in the last couple of electoral cycles, and even then it’s only a very slight effect) – in the north of the country the left does better in cities and the right in rural areas, but this is reversed in the south, largely for reasons of land distribution following the reconquista. Rural areas are over-represented in the Spanish system, but so far as the partisan effect goes, things basically balance out.

Granada, Sevilla, etc. are much more right wing than their rural surroundings. Going on, Madrid is also famed as a right wing city (I think it has always voted to the right of the country, though possible there may be one or two exceptions), whereas Extremadura (the two red provinces in the SW on the map) is very rural.

One thing of particular interest in Spain (though off-topic from the odd / even seats) is the effect of the size of PP/Vox and PSOE/Podemos/Sumar within their blocs. In effect, 2 elections could finish 50-50 between the blocs, but if the intra bloc splits are 30-20 vs 40-10 or 25-25 then that can hugely advantage one side or the other, and there are a bunch of intra-bloc thresholds which make a huge difference. For instance, Sumar won about 12% at the last election: if they’d fallen to 10% they would have lost a huge heap of seats, but in all directions (even if that 2% went directly to the socialists). The small (and variable) district magnitudes combined with the increasingly multi-party system make the exact balance of votes within blocs hugely important, and the effects can seem somewhat random.