How proportional was the 2021 Scottish Parliament election?

The final Scottish Parliament results came in on Saturday night after a roller-coaster two-day count (please let’s never have that again!). The Scottish Parliament was created with a broadly proportional voting system to ensure that seats match votes but how proportional was the election?

The Additional Member System (AMS) ensures that the link between seats and votes is far more representative than Westminster’s First Past the Post (FPTP).

A good voting system should have a strong link between votes and seats. That a party can win a majority on 43% of the vote (2019) – let alone 35% of the vote (2005) – is a clear sign that First Past the Post fails to facilitate proper representative democracy. There are a number of ways to measure how representative an electoral system is but the most widely-known method is the Gallagher Index. There is no need to go in the maths behind it here but a low Gallagher Index score shows good proportionality while a higher one indicates bad proportionality.

The Gallagher Index scores for the previous Scottish elections, according to Electoral Reform Society calculations, are shown below.

1999: 8.5, 2003: 8.2, 2007: 9.1, 2011: 8.6, 2016: 6.2

On their own the Gallagher index scores do not how much but compared to UK election scores we can see that the Scottish Parliament is far fairer than First Past the Post used at Westminster. Take the 2015 election for example, which had the Conservatives win a majority on 37% of the vote and UKIP winning 13% of the vote but only one seat. The Gallagher score on that occasion was 15, with all other elections since 1974 having Gallagher indices with double digits.

The 2016 election was therefore the most proportional since the advent of devolution in 1999 but what about the most recent 2021 election? Upgrade Holyrood’s own 2021 Gallagher index calculations gets a figure of 7.8, making the 2021 election the second most proportional election since 1999. (Update: Ballot Box Scotland has a figure of 7.3 so I will need to revisit my calculations but either way we’re in the same rough area).

This can be seen from the performances of each party: SNP (regional vote share 40.3%, seat share 49.6%), Scottish Conservatives (23.5%, 24%), Scottish Labour (18%, 17.1%), Scottish Greens (8.1%, 6.2%) and the Scottish Lib Dems (5.1%, 3.1%). Overall, this is broadly proportional although the SNP outperform due to their constituency dominance.

As mentioned, the Additional Member System, despite being far more representative in terms of proportionality than First Past the Post, has its flaws. Alba’s plan to game the system, by explicitly calling on voters to vote SNP in constituencies and Alba on the list, shows one clear fault: the risk of so-called satellite parties standing in the regions with an explicit intention of artificially exaggerating the support of one party. Had a supermajority materialised, the Gallagher index would likely have rivalled UK election scores. The same goes with All for Unity which had a similar strategy for unionist voters.

READ MORE: Salmond’s Alba venture exposes Scotland’s voting system flaws

The system may not have been successfully gamed by Alba on this occasion but the party has certainly exposed a key flaw of the system, highlighting the need for reform.

This is compounded by the fact that AMS is far from perfect in other aspects. The ratio of constituency to regional seats creates a constituency bias – a party could win a majority on constituency seats alone on less than half the vote (the SNP were three away from doing so in 2021). The version of AMS in Wales is even worse in this respect with 40 constituency seats and 20 regional ones. Mark Drakeford’s Welsh Labour won 30 seats but on less than 40% of the vote.

Furthermore, there is no direct mechanism to ensure national proportionality. Measures to ensure regional proportionality accumulate to deliver broadly nationally proportional results, however, they stop short of explicitly doing so.

In addition to that, another problem of AMS is the continued existence of FPTP seats which encourages tactical voting. Tactical voting will always happen in any system to some degree, but as many people as possible should be able to vote for their first choice, not their least favourite option.

Lastly, voters have limited choice over candidates within different parties – the list component of AMS ensures better overall representation but the fact that it is closed means that voters get what parties present. There are other flaws too which can be read about here.

It cannot be said enough that AMS is an improvement on FPTP. The UK needs proroportional representation. AMS results are broadly proportional and constituents are represented by a diverse range of parties but that doesn’t mean there are not better alternatives.

A sticking plaster approach to improving on AMS would be to retain the current system and add levelling seats like in Germany to ensure that seats overall match the regional vote. This would improve proportionality but voters are unlikely to support more politicians. An open-list element could be added to the regional list which would ensure voter choice like on Bavaria. In theory this would be an excellent improvement but adding a potential third ballot risks complicating things.

A better alternative would be the Single Transferable Vote, which would involve multi-member constituencies where voters rank candidates by order of preference. STV would deliver proportional outcomes while give voters a significant amount of power over candidates and parties. It is currently supported by the SNP, Lib Dems and Greens, as well as leading reform groups such as the Electoral Reform Society.

There is also the less talked about system of open-list PR with levelling seats, which is used in the likes of Denmark, Iceland and Sweden, and is advocated by Ballot Box Scotland (who has been an invaluable resource during the 2021 election it must be said!). Under such a system there would be multi-member systems where voters choose one party but get to vote for candidates within that party. Once seats are distributed there would then be a mechanism to allocate additional seats to ensure proportional outcomes overall.

Overall, the 2021 Scottish election delivered a broadly proportional outcome, which should be commended. the fact that Westminster still uses First Past the Post is a travesty, putting us at odds with most democracies. That said we should learn still from the flaws exposed from the likes of Alba and All for Unity and reform Scotland’s voting system for the better.