A nerd’s guide to how the six MEP seats will be split

The result of the European Parliament election in June is almost a foregone conclusion in voting terms.

Based on consistent polling data, the Labour Party is expected to obtain an absolute majority of votes with the only factor in play being the size of the voting gap between the major parties.

However, a question mark hangs over the number of seats each of the two major parties will obtain. It is extremely unlikely any other party or independent candidate will be elected, leaving the two major parties battling it out for the six available seats.

Polling data so far suggests the result will give the PL and the PN three seats each, thus returning to the situation of parity in 2014.

However, polls only capture how voters intend to cast their first preference vote, which on its own may not be enough to understand how many seats a party will win.

The Maltese electoral system allows voters to transfer their vote between candidates – even candidates from different parties – by awarding subsequent preferences.
This means the number of votes a party collectively achieves at the first count in the European election does not automatically translate into a proportional number of seats.

To try and understand what might happen on 8 June, we took a look back at the past European elections to find out what happened.

The first two EP elections in 2004 and 2009 elected five MEPs not six since EU rules at the time meant the lowest threshold of seats a country could have, based on its population size, was five not six.

However, midway in the 2009 legislature, the Lisbon Treaty came into force and Malta was awarded an additional seat. This was filled by Labour candidate Joseph Cuschieri, who had been the last man standing at the last count.

In 2014 and 2019, Malta elected six MEPs and will be doing the same in two months’ time.

The threshold: Finding the quota

Malta’s electoral system determines a threshold of votes that a candidate has to achieve to be elected. This is known as the quota. Sometimes a candidate can be elected without reaching the quota if they are the last person on the ballot and still seats to be filled.

In the general election, each district has its own quota but in the EP election all of Malta and Gozo are considered as one large district and so a national quota is set.

The quota for the June election will be determined by dividing the number of valid votes cast by seven (one more than the available seats) plus one. There are mathematical reasons for this formula, which we will not delve into because it goes beyond the scope of this analysis.

The size of the quota thus depends on the number of valid votes cast, which is variable, and the number of seats, which is fixed.

In the 2019 EP election the quota a candidate had to achieve to get elected was 37,174 votes. In the 2004 and 2009 elections, the quota was much higher at 40,954 and 41,362 respectively because the valid votes cast were divided by a smaller number (six) to reflect the five seats that had to be elected.

According to the last MaltaToday survey (March), which projected a turnout of 76.6%, the quota would stand at 40,297 votes. A lower turnout would result in a smaller quota.

The seats: Determining the quota share

To arrive at an approximation of seats the parties could win, we divided the total number of valid votes each obtained on the first count by the quota.

In 2019, the PL’s first count vote translated into a share of 3.80 quotas – three full quotas, which are equivalent to three seats and 0.80 of a quota. The PN’s first count vote translated into a share of 2.65 quotas – two full quotas, equivalent to two seats and 0.65 of a quota.

All other political parties and independent candidates collectively polled 0.54 of a quota on the first count. This means the collective vote of all third parties and independents at the first count was not enough to capture even one seat.

These numbers show that the two major parties could determine that they would be sharing five seats between them (three for the PL and two for the PN), leaving just one seat in play.

By looking at the fraction (0.80 for the PL and 0.65 for the PN) one could argue that in 2019, on the first count the PL was in pole position to obtain its fourth seat since its remaining share was larger than the PN’s. The same happened in 2014 when the six seats were divided equally between the two major parties.

In 2014, the PL had a share of 3.74 quotas against the PN’s 2.80. While it was immediately clear that the PL would win three seats and the PN two, the remaining share suggested the PN enjoyed a slight advantage over Labour – 0.80 against 0.74 – to win its third seat. Eventually, this advantage did translate into a seat for the PN although as we will see further down, luck also played its part.

The same happened in 2004 when the PL had 2.91 quotas against the PN’s 2.39. Each were assured two seats at the first count, leaving in play the fifth and final seat. However, with Labour’s fraction standing at 0.91 and the PN’s at 0.39, the former was in pole position to capture its third seat, as it eventually did.

If we apply this mathematical formula to the MaltaToday March survey results, the PL would end up with 3.70 quotas on the first count and the PN with 2.98 quotas. All the rest would have 0.32 quotas to their name.

This means the PL is assured of three seats and the PN of two. However, if past election patterns are anything to go by, the PN is likely to win its third seat since it currently enjoys a higher fraction than the PL – 0.98 versus 0.70.

Although this straightforward mathematical analysis of first count votes as a share of the quota is a plausible way of determining where Malta’s sixth seat could go, it is not a foregone conclusion. This pattern may still be disrupted.

The inheritance: Influence of vote transfers

The electoral system allows voters to shift their preference between candidates of different parties (which is perfectly legal and does not spoil the ballot) or decide to give a set number of preferences and stop (rendering their vote non-transferable at some stage). These two scenarios can also play a determining factor in the outcome of that sixth seat.

Indeed, by repeating the quota share exercise at every count, one can track the size of the remaining fraction to determine whether the party is getting closer to achieving a full quota or moving away from that possibility.

Our first example is from 2004. Alternattiva Demokratika’s Arnold Cassola had secured more than 22,000 votes on the first count, an impressive performance for a third-party candidate. He singlehandedly obtained 0.56 of a quota on the first count and until his elimination from the race had increased this to 0.71 since he kept inheriting votes at a steady pace from candidates who were eliminated before him.

Roll forward to 2014 and the election result shows us that on Count 1 the PL had 3.74 of a quota and the PN 2.80. This suggested that the PN was in pole position to win its third seat.

However, by Count 27, which was the penultimate count, the PL had increased its quota share to 3.84, while the PN’s stood at 2.82.

At this late stage in the race, the PL’s remaining share (0.84) was larger than the PN’s (0.82) thus indicating that Labour had turned the tables and was now in pole position to capture its fourth seat.

The reasons why this happened are twofold: PL voters were more disciplined than their PN counterparts in voting for all Labour candidates thus ensuring their vote remained within the red block; PL candidates inherited more votes from third party candidates and independents when these were eliminated.

Nonetheless, by a stroke of luck the PN managed to clinch its third seat by a whisker.

At Count 27, the PL’s Joseph Cuschieri was eliminated from the race because he had the least votes. The only remaining candidates where Clint Camilleri, who had 23,273 votes to his name, and Marlene Mizzi (26,242 votes) for the PL, and Therese Comodini Cachia (29,481 votes) for the PN.

Upon elimination, Cuschieri’s 16,613 votes were redistributed according to voter preference. Camilleri received 6,101 votes, Mizzi 9,388 votes and Comodini Cachia 99 votes.

At Count 28, Camilleri’s total stood at 29,374, Mizzi’s at 35,630 and Comodini Cachia’s at 29,580.

This meant that Camilleri, who had the least votes, was eliminated and the last two seats available went to Mizzi and Comodini Cachia in a race that saw the parties elect three MEPs each.

Prior to Camilleri’s elimination, the PL had enough votes within its kitty to elect both him and Mizzi at the expense of Comodini Cachia but the manner by which Cuschieri’s votes were inherited meant that Camilleri missed the bus.

Had Camilleri inherited 207 more votes from Cuschieri (at the expense of Mizzi), he would have overtaken Comodini Cachia at the last count and Labour would have elected both him and Mizzi.

This example shows that although the quota share parties achieve on the first count does give an indication of how seats will be apportioned, this may change as the race progresses… and as 2014 showed, it could even be down to luck.

The future: What can happen in June

Eyes will be set on Roberta Metsola’s performance this June. She is undoubtedly the PN’s frontrunner but it has to be seen whether she could entice lukewarm Nationalists and middle of the road voters to vote for her.

What the PN will be wary of is having a strong vote of sympathy for Metsola that does not translate into subsequent support for other candidates.

If Metsola hoovers up a significant number of first preferences from voters who then shift to other parties or simply stop at her name, the PN’s share of the quota may be strong on Count 1 but could start to dwindle thereafter thus missing out on the third seat.

At this stage, this is all speculation. Polls can only give a snapshot of voter behaviour and what really counts is how each voter marks their ballot sheet on the day.

A higher percentage of vote transfers between candidates from different parties and independents; or a protest vote whereby voters mark their first preference, and possibly their second and stop, could upset the applecart of history.

At this stage, all we can do is wait, see and play around with mathematical calculations to understand the past without knowing whether it will be reflected in the immediate future.