How the new Senedd election voting system works and why tactical voting is key

Dr Keith Darlington

The Senedd election on May 7 will use a new voting system to elect an enlarged group of 96 MSs (Senedd Members). However, recent research shows that over two thirds of Welsh voters do not understand how the new system works.

In this article, I describe how the new voting system works and why tactical voting still works in the new system – voting Plaid Cymru is the best way to stop Reform.

Unlike the previous voting system, which was predominantly based on the Westminster First Past the Post, the new voting system will be based on proportional representation – meaning that members will be elected according to the number of votes a party receives.

This is a fairer system because, under the previous voting system, parties were elected with narrow majorities at a 35% vote share. This cannot happen in the new system as I will show. However, the new system can still facilitate tactical voting, meaning that if a voter wants to prevent Reform from winning, as seemed to be the case in the Caerphilly by-election last year, they would have a better chance of doing so by voting for Plaid Cymru, rather than Labour, who are trailing in third place.

The Old Voting System

The old voting system used a combination of 40 Westminster-style First Past the Post (FPTP) seats and an additional member system with 20 members. This meant that voters were given two ballot papers in the voting booth: one to decide a constituency member (similar to the Westminster General Election) and the other ballot paper to decide the additional member list.

This system was designed to prevent the much derided FPTP system from delivering a permanent one-party dominated system. However, even with the top-up list, it did not prevent the flaws of FPTP and still led to single-party domination on a vote share of less than 40%. Correcting this unfairness is the primary reason for introducing the new system.

Another major flaw of FPTP is that for voters who lived in constituencies that stack up large majorities for one party, such as Labour in the Valleys, many felt their votes were wasted because the outcome was inevitable in safe party seats. This is not so in the new system because every vote counts irrespective of where you live in Wales.

The New D’Hont Proportional Representation System

Voters entering the polling booth for the Welsh Senedd elections on 7th May will now receive one ballot paper and will be required to vote for a single party (or an independent candidate if one is standing) with just one vote. They will not be voting for a person representing a party but a party.

When the polling booths close, votes will then be counted to calculate the number of votes cast for each party. There will no longer be Welsh constituencies, as there were under the old system. Instead, Wales will be split into 16 regions, and each region will return 6 winning candidates to the new Senedd. This means that the Senedd will be electing 6 x 16 = 96 members. This means that each party can field up to 6 candidates and they will be selected according to their position on the list submitted by each party.

Calculating the Winners in each Region using the D’Hont Method

The new system uses a formula called the D’Hont method to calculate the winners in each region. This method delivers winners in proportion to the votes cast. Thus, if a party wins half of the votes, it is likely to win three out of six seats. But the D’Hondt method uses a mathematical formula that calculates exactly who gets each seat. It does it by allocating the first seat to the party with the highest number of votes cast in that region. It means that the top listed candidate in that party will get the first seat.

Each subsequent seat winner is found by repeating the process but with the party vote of the top candidate halved and then seeing which party has the highest votes. Again, the party with the highest total wins the next seat and the candidate in the position on the list will be chosen. The example below shows how this works in practice.

Example of Seat Allocation Using D’Hont

Suppose there are 4 parties standing in a region. Lets assume for simplicity that these parties are called A, B C, and D, and that the votes are counted and give the following results:
Party A = 20,000
Party B = 14,000
Party C = 6,000
Party D = 2,000

The first round winning candidate is Party A because they have the highest vote share of 20,000.

The second round winning candidate is now calculated using the formula. For Party A, their vote is now halved from 20,000 to 10,000. So the largest number from this group now wins the second seat. The numbers are now Party A =10,000, Party B = 14,000, Party C = 6,000 and Party D = 2,000. The largest of these is Party B, and so Party B wins the second seat.

For the third round, because Party B won the last seat, their number is now halved and therefore, the numbers for the third round are Party A 10,000, Party B = 7,500, Party C = 6000, and Party D =2000. The largest of these numbers are for Party A and so they win the third seat.

For the fourth round, again Party A’s vote is now halved again because they won the last seat and therefore the new numbers are Party A =5000, Party B = 7500, Party C = 6000, and Party D = 2000. Of these numbers Party B has the highest and therefore, they win the fourth seat.

For the fifth round, Party B’s vote has now changed to 3725 and all others are the same as before. So, the new totals are Party A = 5000, Party B = 3,725, Party C = 6,000 and Party D = 2000. Of these, Party C now has the highest total and therefore they get the fifth seat.

For the sixth and final seat, only Party C halves to 3,000, otherwise the others the same as the previous round. Therefore, the totals are Party A =5000, Party B =3725, Party C = 3,000, and Party D = 2,000. Of these numbers Party A is the highest and therefore, they win the last and final seat. This means the final seat allocation becomes:
Party A = 3 seats
Party B = 2 seats
Party C = 1 seat
Party D no seats.

Conclusions

These seats are allocated proportionately but Party D gets no seats because their vote is below the calculation threshold. This is one of the criticisms of the system, it works in such a way that it can leave parties with a low vote share with nothing. However, one of the main criticisms of PR in the past has been that small parties can have a disproportionate influence on the results. That is unlikely to happen with D’Hont.

Note also that D’Hont favours the largest parties. This means that tactical voting that was used in the Caerphilly by-election last year to stop Reform still works. In my example shown above if voters wanted to prevent Party B getting seats, then they could lend their votes to Party A. It is clear, from all the opinion polling evidence, that this Senedd election is between Plaid Cymru and Reform and voters who want to prevent Reform from winning should therefore vote Plaid Cymru.