Is STV part of the problem in NI?

Today I attended a seminar at University College London: “Is STV part of the problem in Northern Ireland?” Attended by academics and public officials, eg. NIO and the Electoral Commission. David Ford and I were the only political representatives.

Sydney Elliott applauded the merit of STV in maximising the choice of parties and their and candidates in NI, but mooted whether this fits the “communal need” in NI. In other words, is the abundance of choice assisting a polity that is less inclined to negotiate? More on this latter.

My comment to Sydney Elliott was whether there was any evidence that voters in 2003 were more or less inclined to cross the communal divide in their preferences. I cited the RTE 1998 exit poll that showed that only 15% of the UUP and SDLP voters gave the other *any* secondary preferences, while their transfer rate to Alliance was 35%. Elliott acknowledged that Alliance has traditionally been the party that voters transfer through.

Someone then asked the “so what” question, i.e. are we to blame the STV electoral system, or other factors, for NI election results?

David Farrell gave a very good and concise review of six electoral voting systems used in democracies: Most relevant for today’s discussion are AV and STV. AV is majoritarian: there is one successful candidate per constituency, who requires 50% + 1 to get elected. Under AV, voters vote in order of preference for all the candidates on the ballot paper. As an illustration, the candidate with the least amount of first preference vote is eliminated, and these ballot papers are redistributed to whoever the voters marked as their next preference. This process continues until one candidate receives 50% + 1 of the total votes cast.

STV is proportional: there are more than one candidate per constituency, with each candidate required to reach a quota (determined by a formula: Quota = (total votes cast/number of seats in constituency) + 1. As with AV, voters vote in order of preference for all candidates on the ballot paper, but the redistribution of votes is more complicated. The aim of all proportional voting systems is to translate the percentage of votes to an equivalent percentage of seats available.

Benjamin Reilly was introduced as a protagonist in the debate between STV and AV. The main issue he wanted to address was preferential voting and political moderation. He began by explaining the benign theory that affording voters with more than one preference (vote) encourages moderation by other parties/politicians. In ethnically-divided societies, this means appealing across the communal divide (after shoring up your own first preference vote).

Reilly continued by presenting three aspects of this co-operative theory. The first is a very important if not vital difference in the effect of STV versus the AV voting system. Under AV, thresholds to get elected are high (it’s a majoritarian voting system). As smaller party candidates are less likely to get elected themselves, but larger parties needing their transfers to get their own candidate elected, smaller parties yield *influence* upon the wider political system. For example, a larger winning party will usually concede a particular policy issue in exchange for the smaller party’s secondary support.

However, under STV it is easier for candidates from smaller parties to actually get elected, because the relative threshold for victory is lower. Therefore, smaller parties are less likely to make any pacts with other parties pre-election. Instead, they will use what leverage they have (i.e. relative number of seats) post-election, to participate fully in any coalition government.

In his country comparisons, Reilly mooted whether the AV system was responsible for wiping out moderate parties in Fiji, because it allowed intra-communal competition. In the Q&A session, I rejected the proposition that AV was the cause, per se, because the same happened in NI under STV. I gave the example of how Unionists scrapped STV in the 1929, in order to eliminate dissident Protestants/unionists; the Unionist Party was quite happy with two power blocs: one unionist, one nationalist (see Robin Wilson, “Northern Ireland: What’s Going Wrong”, Institute of Governance, Belfast, December 2003, p. 15). Nevertheless, the NILP and Ulster Liberals survived, but Alliance emerged as their challenge. STV benefitted Alliance, and its relative success as a declared cross-community party put paid to NI parties based on functional (left-right) issues.

There are scenarios where under STV, Alliance could be eliminated immediately, as well as scenarios where STV could assist its proportional strength. If you reduced the number of elected representatives per constituency, e.g. from 6 to 3 MLAs, the quota required to get elected would be so high that no party with less than 25% of the vote would be assured to win. Here, Alliance would be eliminated thoroughly. But perhaps worse, in constituencies where unionists or nationalists received less than 25% of the vote, there is the danger of such minority group not fielding any candidates. Horowitz’s answer to this is to ensure mixed communal constituencies. Yet this places constant burdens on boundary commission officials.

Conversely, by increasing the number of representatives per constituency (which would be appropriately geographically larger/amalgamated), with the quota threshold thus lowered, smaller parties are better assured of some level of electoral representation.

Donald Horowitz argued that there are two broad approaches to accommodating political difference: (a) appeal to moderation, coalescing in a “centre ground”; (b) Co-opt the extremes in an all-inclusive model (the motivation is that the offer of participation will make extremes reluctant to exclude themselves).

Horowitz blames STV for discouraging co-operation among parties in NI, because of the fear/need for NI political moderates to look over their shoulder to their relative extremes.

Horowitz also said that the Agreement contained mechanisms to promote proportionality and all-inclusiveness, as well as mechanisms to attempt to break down divisions. Recalling his two-approaches theory, there are obvious inconsistencies within the Agreement.

The final Q&A raised some very interesting points. David Ford proposed that one cannot ignore the role of communal designations and the lack of collective responsibility (due to the use of d’Hondt to form an involuntary Executive), which may be more significant than applications of STV in getting the MLAs elected to the Assembly.

Another observer said that the majority (of unionists) in NI have rejected the all-inclusive approach, but that it may be too late to return to a centrist approach, because of the electoral weakness of the SDLP (vis-à-vis Sinn Fein). He mooted whether any electoral voting system would make the situation better.

Another stated that majoritarianism [and by implication, the centrist approach] is difficult in an environment of political violence, and perhaps NI is not quite ready to try majoritiarian voting systems.

I wanted to term “majoritarian” defined, and I asked for an STV system with a weighted majority to be considered. This is less than an all-inclusive, rigidly proportional (and involuntary) result, but I would argue not a gross FPTP majoritarian, or an AV-majoritarian, system that requests multi-ethnic constituencies and two-a-side bipolarity with co-operative moderates in the middle.

My final statement of self-interest: why would I opt for a system (i.e. AV-majoritarian) that would relegate Northern Ireland’s most significant cross-community party to an interest group? After all, NI politics is still one of identity conflict.

In essence, all a voting system does is decide the participants for government/Executive formation. The issue of trust, and lack of it, across communities is often cited as a barrier in NI. An AV-majoritarian system actually requires *more* trust among the electorate, in that the voting instructions the voters receive will produce the desired results. With a lack of trust, and STV as the “softer” electoral option, perhaps more concentration needs to be on how to encourage and develop trust once all of the actors have been elected. This is where Ford’s (and others’) points about designations and collective responsibility are particularly relevant.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s