Harbingers of Doom May Be Wrong after All; Philip McGuinness Looks at Voting Figures in Northern Ireland and Casts His Doubts on the Premise That a Nationalist Majority Is Just around the Corner

Article excerpt

Since the 1991 census, the harbingers of demographic doom have had a field day in Northern Ireland.

Forecasts of Catholic/nationalist majorities are being bandied about with triumphalist abandon.

The 1996 and 1997 elections seem to confirm these forecasts. An analysis of the voting figures tells a less dramatic story.

Some facts first: the Catholic share of the population in 1991 was 38.4 per cent. When the "Don't Know/Not Stated" respondents are analysed, what Professor Brendan O'Leary has called the "cultural Catholic" proportion of the population in Northern Ireland has been estimated as between 40 to 43 per cent.

Given the higher birth-rate for Catholics compared to Protestants, the cultural Catholic proportion of the electorage would be about 3 per cent lower, giving a figure for the electorate of 37 to 40 per cent.

Allowing for the five to six years between the census and the three elections, we can revise the figure upwards from 38.5 to 41.5 per cent. For the sake of the following analysis, I will take the midpoint of this range, ie that 40 per cent of the electorate is culturally Catholic.

Taken together, the three elections (1996 Forum, 1997 Westminster, 1997 Local Government) give the following approximate figures: 50 per cent unionist parties, 40 per cent nationalist parties, and 10 per cent middle ground.

Three assumptions have been made. Firstly, the middle ground vote is equally shared between the two traditions.

Secondly, practically all the unionist vote comes from Protestants. Thirdly, the nationalist vote is a Catholic vote.

This gives us a "cultural Catholic" vote of 45 per cent of the electorage.

Thus there is a five per cent discrepancy between the cultural Catholic vote and the cultural Catholic electorage. Such a large gap needs to be explained. One approach is to look at the turnout in elections.

Can it be the case that Catholics are more likely to vote? Yes and no.

Yes in the sense that the constituencies/council electoral areas (what I call "electoral entities") with the highest turnouts tend to have a larger Catholic percentage of the population than those electoral entities along the east coast where turnout is lower.

Such areas have had closely-fought electoral contests for over a hundred years.

No in the sense that very little research has been carried out on differential voting between the two communities within a particular constituency/council electoral area.

One simply cannot draw any conclusions about this situation despite the wealth of anecdote and rumour about postal voting, double voting etc.

I have estimated the effect of this differential turnout on all elections (except European elections) since the mid-Seventies.

This was done by assuming that both communities had an equal turnout within any electoral entity. The percentages of the vote for unionists, nationalists, and middle-ground was applied to the entire electorate of that entity.

This procedure was completed for all such entities for that particular election (17-18 Westminster constituencies; or 100-odd council electoral areas).

Statistics show that when the voting figures are adjusted for differential turnout, the nationalist share of the vote is reduced. …