By Stephen Fisher, 6th May 2021.
There is speculation about how well various new and small parties will do in today’s elections to the Scottish and Welsh Parliaments. Former SNP First-Minister Alex Salmond has recently established the Alba Party, while George Galloway has created Unity 4 All. Both are contesting regional list seats in Scotland. Meanwhile, Abolish the Welsh Assembly has come fourth, with at least 6% of the regional list vote, in all four polls for the Welsh Parliament (Senedd Cymru) since mid-April. Also, Reform-UK, the renamed Brexit Party, is fielding candidates in both countries. These are just some of the various contenders.
This post explains some of the complications in trying to figure out what share of the vote a party needs to get elected in these institutions. It ends up pointing to past experience as a guide and drawing comparison with and the electoral system for the Greater London Assembly (GLA).
TLDR: The experience of all five elections to the Scottish Parliament suggests that around 5.5% of a regional list vote is usually enough to win a seat. It would be rare, but not impossible, to miss out on a seat with 6%. Winning on 5.2% to 5.4% is not uncommon. The lowest D’Hondt ratio ever to yield a seat in Scotland was 4.6% (equivalent to a single party winning a seat on that share). That was in 2003 when the SNP won 5 seats, including 2 list seats, with 23.0% of the Mid Scotland and Fife list vote. For the Senedd, a share of around 6.5% in a region is likely to be enough, especially in North Wales and Mid and West Wales which have historically been more accessible to small parties. By comparison, the GLA has a 5% legal threshold, without which parties would get seats with 3.8% of the vote.
The Scottish Parliament has 129 seats. The electoral system is sometimes described as proportional, but it is significantly different from pure proportional representation. If it were close then a party would win a seat for roughly every 0.8% of the vote, since 100/129=0.8. More precisely, if all the seats were elected by the D’Hondt method of proportional representation a party would be guaranteed a seat with more than 100/(1+Number of seats) per cent of the vote, that is 100/130 = 0.77% of the vote. As we shall see that is much lower than what is actually required to get elected.
The first and most significant deviation from proportionality as it affects the chances of small parties getting representation is the division of Scotland into 8 regions, with 15, 16, or 17 seats each. If all the seats in each region (including both constituency and regional list seats) were elected proportionately then a party would need 6.3%, 5.9% or 5.6%, respectively, to be sure of a seat.
It would be possible to win with much less than this number if there are lots of parties and votes divided evenly between them. Imagine 98 parties getting 1% each and one party getting 2%, for example. The theoretical minimum threshold for representation depends on such unrealistic scenarios.
Within each region only 7 seats are elected by D’Hondt from the regional list, while the rest are elected by first-past-the-post as for Westminster. One way of thinking about the regional list seats is as top-up seats that make the result more proportional after taking into account the outcomes in the constituencies. But since there are more first-past-the-post seats than regional list seats in each region, there is not necessarily enough regional list seats to correct any disproportionality if a big party wins a bigger share of first-past-the-post seats than their share of the vote. That has been quite common as Labour won the lion’s share of first past-the-post seats in several regions in the first three Scottish Parliament elections (1999, 2003, and 2007), while the SNP did the same at the last elections in 2016.
If the regional list seat allocation starts with a big party having already grabbed a bigger share of the constituency seats than their share of the regional list vote then it is harder for the small parties to win seats. The list allocation it becomes a competition between parties that did not collectively win 100%, but the threshold for representation is still higher than it would be if no party won a bigger share of constituencies than their share of list votes.
For example, in Glasgow in 2016 the SNP won all 10 first past the post seats, 10/17=59% of the total, but won only 45% of the regional list vote. Being already over-represented, there was no chance of the SNP winning a regional list seat. How did this affect the threshold for the other parties? They were contesting 7 list seats with 100-45 = 55% of the vote between them, which meant that roughly 55/8 = 6.9% would guarantee a seat. That threshold is up from the 5.6% that would have been needed had no party won a bigger share of constituencies than their share of list votes. That is based on the observation that if no party wins a bigger share of constituencies than their share of list votes then the seat allocation will be the same as if all the seats in the region (including those actually allocated to constituencies) were allocated by D’Hondt using the list vote.
Glasgow in 2016, while not unusual, is apparently one of the more striking examples of over-representation of a large party. I say “apparently” because the lack of enough regional list seats to ensure proportionality generates a strategic incentive to desert parties on the regional list vote if they are likely to win disproportionately many constituencies. That is something the Greens may have benefitted from in 2016 and something that the new Alba party is explicitly trying to exploit when it says it wants to generate a “super-majority” for independence.
In the event, there were enough small parties on regional lists in 2016 for the Greens to pick up a one seat in the West of Scotland on just 5.3% and two seats in Lothian on 10.6% (effectively 5.3% each). In other regions they failed to pick up seats with 4.7% and 4.9%.
The Greens in 2016 seem to have been fortunate but not unusually lucky to pick up three seats effectively on 5.3% each. In 2007 Lib Dems and Greens both won seats on 5.2% of a regional list vote, in Central Scotland and Glasgow respectively. However, in 2011, the Liberal Democrats failed to get a seat on the Lothian list despite winning 5.5%.
As far as I know, the most list seats allocated on the smallest average shares were in 2003 when the SNP won 5 seats, including 2 list seats, with 23.0% of the Mid Scotland and Fife list vote, effectively just 4.6% for their final list seat.
Overall, the experience of all five elections to the Scottish Parliament suggests that around 5.5% of a regional list vote is usually enough to win a seat. It would be rare, but not impossible, to miss out on a seat with 6%.
The electoral system for the Senedd is structurally the same as that for the Scottish Parliament, but the numbers are smaller, making it harder for smaller parties to gain representation.
This starts with the overall number of seats. With just 60 seats, even with nationwide D’Hondt the threshold to guarantee representation would be 1.6%. Wales is divided into 5 regions with 11, 12, or 13 seats per region. As in Scotland the regional split raises the bar for smaller parties, to 7.1%, 7.7% or 8.3% if all the seats in each region (including those actually allocated to constituencies) were elected by D’Hondt.
Whereas in Scotland the ratios of regional list to constituency seats are 7/8, 7/9 or 7/10 depending on the region, in Wales, with 4 top up seats in each region, the ratios are much lower at 4/7, 4/8 and 4/9. The lower the ratio the less chance the regional list votes have of correcting disproportionalities from the constituency results. The highest ratio in Wales is, at 0.57, still considerably lower than the lowest for Scotland, which is 0.7. This means that the barriers for smaller parties are greater in Wales as a result of the fewer list seats.
In the three South Wales regions Labour typically wins all or nearly all of the constituencies, and never enough votes to win any regional list seats. To take the most striking example from last time, Labour won all the seats in South Wales West and 39.5% of the regional vote. That means the other parties, with 60.5% of the vote between them, were competing for the 4 list seats. That means they needed 12.1% to be sure of a seat. In the event, however, Plaid Cymru picked up a second seat in the region on a D’Hondt ratio of 8.6%.
Also in 2016, Plaid Cymru won a fourth seat in Mid & West Wales on 6.6%, and Labour picked one up 6.5%. UKIP got a second seat on 6.2% in North Wales. Some of these well-below-threshold successes last time were helped by the fact that other parties secured 9.4% overall on the regional list vote. But this didn’t seem to help the minor parties in the South secure seats on less than 7% of the vote.
In 2011, votes in the South were more widely spread across more parties than in 2016. As a result, in South Wales East the Conservatives won a third seat on a D’Hondt ratio of 6.5% and Plaid Cymru won their 2nd seat on 6.0%. The Liberal Democrats won a seat in South Wales West on 6.9%. Meanwhile, final seats in North Wales and Mid and West Wales were again accessible to parties with more than 6% of the regional list vote. In 2007, Plaid Cymru got their final seat in three regions on D’Hondt ratios of 6.2%, 6.4% and 6.8%. In 2003, just two seats were won with D’Hondt ratios of under 7, both by Conservatives. In 1999 none were, because of the very small share of the regional list vote for minor parties, just 5.1%.
If Senedd elections continue to see a relatively large number of small parties attracting a sizeable share of the regional list vote between them, as polls suggest they will today, then a share of around 6.5% in a region has a fair chance of yielding a regional list seat, especially in North Wales and Mid and West Wales which have historically been more accessible to small parties. But this is far from guaranteed.
The Greater London Assembly (GLA) electoral system is simpler than those of Scotland and Wales because there is no regional split, but it is ultimately similar in having constituencies elected by first past the post, and then a set of top up seats allocated by D’Hondt using the London-wide list vote. If all 25 seats were allocated by London-wide D’Hondt the threshold for guaranteed representation would be 3.8%. That is relatively high because there are so few seats in total.
Actually, only 11 seats are allocated by D’Hondt, taking into account the results of 14 first-past-the-post constituencies. Only once since the first Assembly elections in 2000 did a party fail to pick up a list seat because they won too constituencies. Given how dominant Labour now are in the city, it is strange to be reminded that the one exception was the Conservatives in 2004. They won 9 constituencies then, the highest number on record and only once matched (by Labour in 2016).
Since having 14 of the 25 GLA seats as constituency seats has only once in five elections been a source of disproportionality, parties could have expected to win representation with just 3.8% of the vote were it not for the legally imposed 5% threshold. That legal threshold stopped the BNP getting a seat with 4.8% in 2004 and stopped UKIP getting a seat with 4.5% in 2012. Whether it will influence the outcome of today’s elections remains to be seen.