The May 5th Assembly election is less than 5 weeks away. This a review of previous elections since 2003. I have looked at the raw vote for unionist, nationalist and nonsectarian parties and have plotted the changes over time in the various elections.
I have also calculated voter turnout from the Catholic and Protestant voting age population for each election.
If you look at the first set of data the religious background of the voting age population is shown for the 2001 and 2011 census. The 2001 census is relatively straightforward in that the group of None/Other was only 2% of the voting age population and probably only 1% of actual voters.
Polls have consistently shown that the None/Other group vote at a much lower rate than those from the Protestant and Catholic community background. The 2011 census is more complicated. There were 60,000 more foreign nationals in the 2011 census compared to 2001.
The Protestant voting age population increased by 9,000 but this was entirely due to foreign nationals (Eastern Orthodox and Protestants from other immigrant nationalities). The Catholic voting age population increased by 93,000 but 30,000 of this increase was due to foreign nationals. There was also a large increase in the Other/None group to 71,000 but over 20,000 of those were also foreign nationals.
The potential unionist electorate did increase from 2001 to 2011 since some of the None/Other group would vote or transfer to unionist candidates but this increased electorate probably plateaued by 2005.
Extrapolating the addition of new Protestant voters minus emigration and the death of older voters indicates the potential unionist electorate has been stable since 2005. The potential nationalist electorate has increased dramatically between 2001 and 2016. The 2nd chart shows the percentage turnout from the potential unionist and nationalist voter pool. Now it is true that there has been a large increase in the vote for the nonsectarian parties but these voters do have 2nd preferences and I looked at the transfer pattern to estimate the turnout from the unionist and nationalist voter pool.
When you look at the vote for unionist, nationalist and nonsectarian parties there are several notable findings. With a fixed potential unionist electorate the vote for unionist parties in 2015 is essentially the same as that in the 2003 Assembly and 2005 Westminster elections. It did decrease in the 2010 and 2011 elections but this is entirely due to scandals within the DUP at that time. It has since completely recovered in 2015. You will note that there was no decline in the unionist vote when comparing the 2011 Assembly and 2014 Euro election but a massive decline of 43,000 votes in the nationalist vote between those elections.
I have adjusted the percentage turnout from the potential nationalist electorate in each election to account for the increase in the nationalist electorate over time (the unadjusted numbers are based on the 2001 census for the 2003 to 2007 elections). The graph indicates that the nationalist turnout was 3% to 5% higher than the unionist turnout in the 2003 and 2005 elections. By the 2010 and 2011 elections it was slightly less than the unionist turnout. In the 2014 and 2015 elections it was 6% and 9% less than the unionist turnout respectively. The unionist turnout in 2015 was slightly higher than the unionist turnout in 2003 and 2005 but the nationalist turnout declined from 59% in 2005 to 46% in 2015.
While unionist voters are still voting at the same rate as they always have, 20% of the potential nationalist electorate have stopped voting.
It is a myth that there is voting apathy among unionist or nonsectarian voters. The voter apathy is entirely from nationalist voters.
What are the implications for the 2016 Assembly election?
I expect that the unionist electorate will continue to vote at the usual 55% rate. The lower turnout in 2010 and 2011 were aberrations due to temporary scandals.
What will happen to the nationalist voter turnout? Are there any factors that could increase that turnout? The actual 100th anniversary of the Easter Rising will be less than 2 weeks before the election but most of the anniversary events have already occurred. While patriotism could be a factor I expect this effect to be minimal. The EU referendum on Brexit will be in June and the proximity to the Assembly election may affect turnout. There have been 2 Lucid Talk polls on Brexit and they showed that only 10% of SF, SDLP, PBP, Alliance and Green voters want to leave the EU.
The polls also showed that only 20% of unionist party voters want to remain in the EU. It is possible that nationalist voters who own or work for businesses that have significant trade with EU countries may be more motivated to turn out to vote for pro EU parties in the Assembly elections but it is not clear if this will be more than minimal.
There is one group of voters that should be highly motivated to vote for pro EU parties and that group are the EU nationals that live in Northern Ireland. Their status in Northern Ireland could be jeopardized by Brexit. They cannot vote in the Brexit referendum but are allowed to vote in the Assembly elections.
The Department of Social Development recently released some data on foreign nationals. As of July 2015 there were 145,000 Non UK nationals living in Northern Ireland with a national insurance number (NiNo). Almost 30,000 were from the Republic of Ireland. The other 115,000 included approximately 85,000 from European countries. That 115,000 does not include ethnic nationals who were born in Northern Ireland (probably only 2,000 or so) or those who have arrived since July 2015. I estimate 90,000 EU nationals of voting age living in Northern Ireland as of April 2016. Only 30,000 are registered to vote.
Surprisingly there appears to have been little interest by the pro EU parties in registering those 60,000 unregistered prior to the April 18th deadline for voter registration. They generally live in the same streets where the registered foreign nationals live. There are 3,000 unregistered EU nationals in East Belfast and one would think that the Alliance Party would register these so they could win a 3rd seat from the DUP. There are 2,000 unregistered EU nationals in North Antrim and it is surprising that the SDLP are not registering them considering that they lost the seat in 2011 by only 600 votes. There are 1,000 unregistered in Strangford and the SDLP lost by only 450 votes in 2011. Similarly for Sinn Fein they lost a seat in Upper Bann by only 400 votes to the UUP and there are over 3,000 unregistered EU nationals in that constituency. In Fermanagh South Tyrone Sinn Fein could lose a seat to the SDLP based on the 2014 council results but with 4,000 EU nationals not registered to vote they could take a seat from the DUP to cancel out that loss if even half that number were registered and voted.
In summary, I expect a continued high turnout of voters from the unionist voting bloc. There may be a slight increase in native nationalist voter turnout related to the 100th anniversary of the Easter Rising and the Brexit referendum. There could be a significant increase in voter turnout among EU nationals concerned about Brexit but this will probably be limited due to the low voter registration rates among those communities.