Transcending an Ethnic Party System? The Impact of Consociational Governance on Electoral Dynamics and the Party System
by Paul Mitchell
From: Aspects of the Belfast Agreement edited by Rick Wilford (Oxford University Press 2001)
Could a yes-no institutional cleavage emerge in the Assembly.
Given the consociational rules on which the internal aspects of the Belfast Agreement are premised, devolution can only progress by means of concurrent consent within the nationalist and unionists blocs in the Assembly. The 'higher' threshold -'parallel consent' - requires absolute majorities within both blocs. Assuming all 108 members vote, this means that a minimum of 22 nationalists and 29 unionists are required to pass ' key decisions'. The less onerous 'weighted majority' rule - requiring at least 40 per cent of both blocs, plus 60 per cent overall - means for example, that if all 42 nationalists voted for a measure, the minimum number of unionists required would be 24.
To date the practical political focus has been on competitive dynamics within unionism, not least because all 48 nationalist members of the legislative assembly (MLAs ) are obstensibly 'Yes' voters. David Trimble's party has 28 members, although one, Peter Weir, resigned the party whip and now sits alongside the DUP in the Assembly. Thus, assuming for the moment that only Weir is a lost cause, Trimble has 27 votes plus the two provided by the PUP, giving him 29 'Yes-unionist' MLAs , the bare minimum need to win a parallel consent division in the Assembly. Clearly the UUP leader can not afford any further defections if the Assembly is to be workable. The other side of this sharp division within unionism is that the 'No' unionists are within striking distance of reaching the de facto 'minority veto' provision of the consociational architecture - the 'petition of concern' - which requires the signatures of 30 members to trigger the special voting procedures. They currently have 28 votes, plus most probably Peter Weir, placing them only one defection short of the ability to trigger the procedures. (34)
A key question is the extent to which the new political institutions may ameliorate the previously intense communal cleavage by superimposing a new 'pro power sharing versus anti power sharing division'. Part of the explanation is that the Agreement cannot work, without an informal coalition government among at least the UUP, SDLP and Sinn Fein, and that this mutual reliance may foster some implicit electoral co-operation as these parties transfer to other 'Yes' candidates. Preferential electoral systems such as STV allow what Horowitz 35) has called 'vote pooling', an incentive towards some inter-ethnic accommodation, facilitated by making parties to some extent dependent on other ethnic groups for vital transfers (36). Evans and O'Leary (37), using survey evidence, found some limited but significant evidence of such effects - see also O'Leary, in Chapter 4 of this volume. They found that about 9 per cent of voters did transfer from nationalist to unionist and vice versa and not surprisingly the willingness to cross the communal divide was greatest within the 'Yes' camp. This is significant given that the 1998 contest was the first election after the Agreement and because there were no explicit transfer pacts among the 'Yes' parties (38). If the Assembly survives long enough for a second election - - feat no NI regional parliament has managed since the old Stormont regime - these intercommunal co-operative electoral practises can be expected to become more important.
Finally, the UUP's failure to trop the first preference vote was mitigated
and somewhat masked by its emergence as the largest Assembly party. The
UUP enjoyed a very significant seat bonus of 4.6 per cent, due to transfers
made possible by the return to STV (39). As noted above, some of these transfers
came from the SDLP. However, Evans and O'Leary (40) found that many of the
transfers to 'Yes' unionists were from 'No' unionist candidates. This suggests
an interesting tension between two forms of electoral rationality: voting
'sincerely' according to one's communal preference - 'No' unionists preferring
to transfer to other unionists, even 'Yes' unionists, than to anybody else;
and tactical rational voting dictated by the institutional rules of the
Assembly for'No' unionists seeking to immobilise the Assembly, more 'Yes'
nationalists are better if this means fewer 'Yes' unionists. (41). It seems
likely given the history of communal polarisation that 'No' unionist voters
'used their preference schedules to say they preferred "yes but sceptical"
unionist candidates to nationalist candidates of whatever kind (42). When
faced with situations in which there are no other 'No' unionist candidates
available, a DUP voter for example, would need to 'transcend' powerful communal
loyalties in order to 'rationally' transfer to a nationalist candidate rather
than to a 'Yes' unionist in order to help immobilise the Assembly. The next
section examines actual transfer patterns. |
