As this week’s UK-wide referendum approaches on the question of changing the voting system for parliamentary elections, Catholics may be wondering whether the Church’s teaching offers any guidance on how to vote. Peter Scally SJ explores the question.
Back in February, it was reported that a group of ten Church of England bishops declared their support for adopting the Alternative Vote (AV) in UK Parliamentary elections, for ‘ethical and moral reasons’. One of their number, Rt Rev Michael Langrish, the Bishop of Exeter, said he backed the change ‘on the grounds of justice and accountability,’ whilst another, Rt Rev Alan Wilson, the Bishop of Buckingham, put it in terms of ‘truth’.
That they choose to do this together, as bishops – and use such language – suggests that Christian faith and Christian ethics, in their opinion at least, have some bearing on the question. But do they? And if they do, is there anything in the teaching of the Catholic Church that Catholic voters, and perhaps other Christians, might find useful when trying to inform themselves on this issue?
The Catholic Church has a considerable body of what is called Social Teaching, contained in its most explicit form in the documents of the Second Vatican Council and certain Synods of Bishops, as well as in the encyclical letters of a succession of popes, dating back to Pope Leo XIII’s Rerum novarum in 1891. Of course, democracy as we know it today was by no means widely established in Pope Leo’s time. The question of democracy itself is perhaps most closely examined in Centesimus annus, the encyclical Pope John Paul II wrote on the centenary of Rerum novarum, in 1991. Those looking for guidance on the question of the merits of different voting systems may be disappointed to read:
The Church respects the legitimate autonomy of the democratic orderand is not entitled to express preferences for this or that institutional or constitutional solution. Her contribution to the political order is precisely her vision of the dignity of the person revealed in all its fullness in the mystery of the Incarnate Word. 
This would seem to rule out the idea of our finding a simple ‘yes’ or ‘no’ to the question in hand, but we should not lose all hope of finding guidance, because the preceding paragraph states:
The Church values the democratic system inasmuch as it ensures the participation of citizens in making political choices, guarantees to the governed the possibility both of electing and holding accountable those who govern them, and of replacing them through peaceful means when appropriate. Thus she cannot encourage the formation of narrow ruling groups which usurp the power of the State for individual interests or for ideological ends.
That one paragraph offers us, at least, some principles – some criteria which we might apply when considering the merits of different voting systems. Two of the principles are stated quite explicitly here: participation and accountability. A third is suggested by those final words about the undesirability of narrow ruling groups usurping power for their own interests – the principle of the common good, which is found throughout Catholic social teaching.
So the Church’s teaching would encourage us to examine the merits of the Alternative Vote in these very specific ways: Will it favour the active participation of voters in making political choices? Will it increase accountability – in particular voters’ ability to replace their representative or their government when appropriate? Will it lead to better government in the interests of the common good, as opposed to allowing the interests of a minority to dominate?
Before we can get very far in answering those questions however, it will help to look first at what would be the likely effect of implementing AV in practical, political terms, in the UK.
This is not an exact science. Given that there has never been a General Election using AV in the UK, it is surprising how confidently people will tell you that the result of some past General Election, ‘had it been held under AV, would have been such-and-such’. They are claiming more than they can possibly know. Voters’ second and third preferences were not recorded – nor was there much interest in them in the past, even on the part of pollsters – so we simply cannot know what they were or would have been.
More recently, however – since AV has emerged as a real possibility – there has been opinion polling on hypothetical second and subsequent preferences, so we can get some idea of the effect it would be likely to have on election results.
If the question is asked, ‘which party or parties will benefit?’ one would expect AV primarily to benefit the party or parties that are in a position to attract second preference votes. Out of the three largest UK-wide parties (let’s imagine for a moment a seat where only these three parties have fielded candidates) that would mean the Liberal Democrats, since they are perceived to be in the centre ground and therefore likely to be the second preference of both Conservative and Labour voters.
It is difficult to predict what difference AV would make to the nationalist parties in Scotland and Wales. Among the smaller parties, one might expect the Greens to attract some second preferences, though not necessarily enough to win them any more seats, and UKIP to attract some Conservative second preferences. It would also be possible that both of these parties attract more first preference votes than before from supporters who were hitherto reluctant to ‘waste’ their vote. This might also apply to the BNP, but contrary to what some No campaigners are claiming, the BNP would be unlikely to benefit greatly, since supporters of other parties tend to be strongly opposed to them – and AV favours candidates who are ‘broadly acceptable’, those to whom most voters do not strongly object.
The broad thrust of these intuitions is borne out by a survey conducted during last year’s general election campaign, in which 17,000 participants filled in AV-style ballot papers. The projected results by AV, compared to those of the real election, were that the Liberal Democrats gained 30 extra seats while the Conservatives ended up with 20 fewer seats and Labour ten fewer. Another survey by researchers at Warwick University gave the Lib Dems a more modest gain of just ten seats.
These figures, of course, are just estimates and only hold for that one election, but the general picture emerges that, unless the Liberal Democrats are polling badly, then in terms of the number of seats in parliament, they will be the main beneficiaries of the Alternative Vote.
Now we can turn to the three criteria outlined above:
‘The Church values the democratic system inasmuch as it ensures the participation of citizens in making political choices.’ Proponents of AV put forward greater participation – the prospect of a higher turnout at elections – as one of its advantages. The Yes to Fairer Votes campaign quotes this passage from the report of the commission chaired by Roy Jenkins that reported on the voting system in 1998:
Under our system, AV would have a number of positive features which persuade a majority of us that it would be superior to FPTP as a method of choosing constituency representatives. First, there will be many fewer ‘wasted votes’ in the constituency side of the election, and far more voters will potentially influence the result. This, we hope, will encourage turn-out and participation.
The commission is right that, under AV, we can expect more voters to have a sense of being able to influence the result in their own constituency, so the ‘hope’ rather weakly expressed here that this will encourage turnout and participation seems a reasonable one. There are, however, other factors that are reported to affect turnout. Firstly, turnout tends to rise when an election is perceived to be close – when the outcome is not a foregone conclusion, either at the local constituency level or the national level – because, again, voters have a greater expectation of being able to influence the result. Secondly, turnout tends to rise when there is perceived to be a real difference between the parties, so that voters have a sense that it really matters ‘which side wins’ and it will make a difference to their lives. Turnout falls when people can see little difference between what the parties are offering and when it won’t make any difference to them which party ends up in government.
So, what difference will AV make to the closeness of elections? Let us look first at the question of closeness of contests at the local constituency level. One of the claims of the Yes to AV campaign is that it will tackle the problem of safe seats (where turnout is generally lower). Their campaign material says, ‘Too many MPs have their “safe seats” for life. Force complacent politicians to sit up and listen, and reach out to the communities they seek to represent.’ This claim is based on the fact that a candidate will need at least 50% of the votes to win. But there are plenty of seats where the winner does get at least 50% of the votes under the current system. At the last election, there were 218 of these – a third of all seats. It cannot be assumed, of course, that they would get the same number of first preference votes under AV: they could end up with fewer votes in total, but they might end up with more. A popular candidate will be likely to pick up second and third preferences in addition to their first preference votes and amass an even more overwhelming total than before. And there is a party political factor again here – from what we have noted above, it can be expected that AV will, on the whole, make existing Lib Dem seats safer.
So AV is not guaranteed to reduce significantly the number of safe seats, and certainly not to eliminate them. The fact is, of course, it would be undemocratic to eliminate them. If an overwhelming majority of people in a particular constituency support a particular candidate, it is right and proper that he or she should win by a large margin and hold what will be regarded as a ‘safe seat’ – that’s democracy!
Secondly, though, what effect would AV have on the closeness of elections at the national level? Would it help prevent the scenario where the outcome of an election seems a foregone conclusion?
Studies by John Curtice of the University of Strathclyde and the British Election Study suggest two marginal effects: One is that, in terms of seats in parliament, AV tends to exaggerate the effect of landslides a little more even than the current system does. So, in those situations, AV would not make elections closer at national level. The other effect, in non-landslide situations, as noted before, is that by slightly boosting the seats of centre parties – in our case the Lib Dems – it makes a majority government a little less likely, which, it could be argued, is a ‘close result’ and desirable in terms of its effect on voter turnout and participation.
There is, however, that other motivating factor to consider – the sense of there being a difference between the parties, and of it ‘making a difference to my life’ which party wins. Opponents of AV argue that making coalition governments more likely will decrease people’s motivation to vote since, in the compromises necessary to form coalition governments, the differences between the parties are softened and blurred. They argue that it is the power to choose the government that most motivates voters, and that in the making of coalition deals, this power is taken out of the hands of the voters and given to party leaders and their negotiating teams. This picture of the ‘disempowered’ voter is perhaps more a critique of proportional systems such as those used in Belgium and Ireland than of AV, which only makes coalitions slightly more likely in certain circumstances.
On balance then, the claim that AV would lead to greater participation on the part of voters seems questionable. This is backed up by evidence from Australia and some Canadian provinces at the time of the introduction of AV, which shows no consistent impact, positive or negative, on voter turnout.
What difference will AV make to accountability? Will it enhance the guarantee to ‘the governed’ of ‘the possibility both of electing and holding accountable those who govern them, and of replacing them through peaceful means when appropriate’? 
This, again, needs to be considered both at the constituency level (the possibility of electing, holding accountable and replacing an MP) and at the national level (the possibility of electing, holding accountable and replacing a government).
Under the current system, it is already, of course, possible for voters to remove their MP. In the last General Election, for instance, former Home Secretary, Jacqui Smith, after embarrassing revelations about her expenses claims, was ousted from her Redditch seat on a swing against her of almost twice the national average.
The main difference AV would make at the local level is that a candidate needs broader support to get elected, although some of that support can be relatively weak. Under the current system, a candidate can quite easily be elected with a minority of the votes – in fact most are – and in the last General Election, 109 MPs were elected with less than 40% of the vote.
Proponents of AV say it will mean an MP only gets elected if they have the support of at least 50% of the voters, but it is arguable whether putting a candidate as your second, third or fourth preference constitutes your ‘support’ in the same sense that a first preference vote does. They also argue that the current system is poor on accountability, because there are circumstances when an MP could survive despite, say, 60% of the voters being against him or her. But if that 60% do not agree on who the MP should be (for instance if they are split evenly between two other candidates, so that each of these has 30% support, while the incumbent has 40%) then it is not immediately clear that the incumbent should be replaced by either of them for the sake of accountability. That, of course, is precisely why the AV system was devised – in order for second and third preferences to be taken into account to decide who is elected in a situation like this when nobody attracts a majority of the first preferences. The result with AV, though, will tend to be that the most broadly acceptable candidate, or the least objectionable candidate, prevails. Whether this is a genuine step forward for accountability is questionable – in effect it will make it a bit easier to remove a Labour or Conservative MP, but probably a little more difficult to remove a Lib Dem one.
AV’s supporters argue further that it increases accountability, because under the current system we have the scenario of people having to vote for someone they don’t really want in order to keep out someone else whom they consider worse – ‘tactical voting’. For instance, the Lib Dems in South Dorset proclaimed on their web site at the last General Election, Vote Lib Dem to keep the Tories out – a clear appeal for tactical voting from Labour supporters. AV eliminates the need for that kind of tactical voting, but leaves those voters ultimately in the same position – Labour supporters in South Dorset might then vote Labour as first preference and Lib Dem as second, and at least have the chance of ‘keeping the Tories out’. Yet, even if they are successful, they are still left with an MP they don’t really want, but whom – AV campaigners would now claim – they ‘voted for’.
Furthermore, AV introduces the possibility of a new and more sophisticated kind of tactical voting. Take, for instance, a seat held by the Conservatives with about 40-45% of the vote, where Labour and the Lib Dems are vying for second place. Let us suppose that in this right-leaning constituency, Lib Dem voters are more inclined to give their second preference to the Conservatives than to Labour. The Conservative candidate’s big fear, then, is that Labour will be eliminated in the first round and Labour voters’ second preferences carry the Lib Dem candidate to victory. The Conservative candidate will want Labour to survive the first round, and the Lib Dem to be eliminated, because then the Lib Dem voters’ second preferences will give an easy victory to the Conservative. Consequently, it would be in the Conservatives’ interest, though difficult to manage, to get a limited number of their supporters to give their first preference vote to Labour.
The Yes to Fairer Votes campaign in Birmingham calls this idea ‘ludicrous beyond belief’ and ‘such a ridiculous hyperthetical [sic] scenario it doesn’t even bear thinking about.’ But one only has to look at the recent election in Ireland (under a different system, known as the Single Transferable Vote) to see how sophisticated vote management and ‘working the system’ inevitably come into play when the winning and losing of seats is at stake.
Probably the biggest obstacle to the personal accountability of MPs, though, is voters’ party loyalty – I may not want this individual to be my representative in parliament, but I vote for him/her because he/she is the candidate of my party, and what I care most about is that my party wins the election. For Catholics and other Christian voters, this may be a real concern when it comes to questions concerning the sanctity of life and the dignity of the human person (euthanasia, assisted suicide, stem-cell research and abortion, for instance) which are also an important part of Catholic Social Teaching, but one that is not followed by most British politicians. AV does not solve this problem, which can only be overcome by some degree of separation between party and candidate – for instance the system used for elections to the Scottish Parliament, where voters have one vote for an individual candidate in their constituency and a second vote for a party in their region, or the Irish system of multi-seat constituencies where the main parties put up more than one candidate in the same constituency. However, AV would allow, for instance, a Labour-supporter with ‘pro-life’ views in a seat with a ‘pro-choice’ Labour candidate to register a first preference vote for an explicitly ‘pro-life’ candidate who stands little chance of being elected, without damaging the Labour candidate’s chances of eventual success.
What, though, would AV do for the accountability of governments? Would it make it easier or harder for the voters to remove a government they don’t like at a General Election?
The truth is, I think, that it would make little difference. As we have seen, it is likely to make it a little harder for one party to get an overall majority (unless that party is in landslide territory, in which case it is likely to make the landslide even bigger). So it would probably make it marginally easier to throw out a one-party government whose popularity is dwindling, but it could make it more difficult to throw out a coalition government.
The small boost it would give the Lib Dems, which makes coalition governments a little more likely, would be interpreted by the No camp as bad for accountability. Looking at the last General Election, they could argue that the voters wanted to throw out the Labour government, but that according to the survey mentioned above, AV would probably have given Labour and the Lib Dems enough seats to form a Lib-Lab coalition – ‘thwarting the will of the people’ to remove Labour from power. A Conservative-Liberal coalition would still have been a possibility, of course, but that crucial political choice would have been made by Nick Clegg, not by the voters. Furthermore, when it is part of a coalition, a party cannot be held to its manifesto pledges, since policy has to be arrived at by compromise – as we have seen in the Lib Dems’ abandonment of the stance on tuition fees which won them so many student votes. All of this, they would argue, militates against the kind of accountability the Church teaches us to desire from a democratic system.
The Common Good
Would the introduction of AV favour government in the interests of the common good and reduce the chances that narrow, sectional groups might ‘usurp the power of the State for individual interests or for ideological ends?’ 
There seems in principle to be some prospect that it would. Since candidates would require broader support to get elected – even if that is only the weak ‘support’ of a second or third preference – they would find it harder to survive by representing the ‘substantial minority’ that is enough to keep them in power under the current system. So AV should limit the chances of extreme, partisan or polarising candidates, and deter the kind of ‘class war’ politics in which a candidate appeals only to the wealthy or only to the poor.
But this effect will not apply everywhere – not, for instance, in constituencies (which will still exist) where the electorate is overwhelmingly drawn from one socio-economic group. Candidates will not need to ‘reach out’ to all sections of society to secure victory in the Conservative bastions of North Yorkshire, Surrey and rural Hertfordshire, nor in the Labour heartlands in London’s East End, Tyneside and Lanarkshire. And even where it does apply, the effect will be limited: most MPs won’t need much more support than they have now – given that a second or third preference vote may end up counting as ‘support’ – to hang on to their seats. So it would be too much to expect the introduction of AV to end the parliamentary culture described by some as ‘adversarial’ and ‘bipolar’ and usher in a new era of consensual politics.
In any case, the politics of ‘class war’ is already in decline even under our current system. Few observers regard the Labour Party as representing working class interests any more, although it is arguable that the Conservative Party has not altogether thrown off its upper class image, especially given the proportion of wealthy and privately-educated members in the present cabinet. Both parties, though, have realised for some years that the key to success – even under the current system – is to appeal to voters beyond their ‘traditional base’. That is what the New Labour project was all about, and also the thinking behind David Cameron’s ‘rebranding’ of the Conservatives. Good, old-fashioned, hard, electoral reality means that the need to ‘reach out’ has already had – and continues to have – a huge impact on British party politics without requiring any encouragement from AV.
Voting and ‘disordered individualism’
There is, however, one more aspect of the Church’s teaching which may have a bearing on the question of electoral reform. It is well known that the Church teaches respect for the human person and the value of the individual, but in recent years, successive Pontiffs have observed – and warned against – a culture of excessive individualism that is developing in Western societies. Put bluntly, this is the ‘me’ culture, in which I am encouraged to think and behave as if the world revolved around me, my needs and my concerns, and to accept nothing as having greater authority in my life than my own views, feelings and experiences. Put more subtly:
Without a renewed education in solidarity, an overemphasis of equality can give rise to an individualism in which each one claims his own rights without wishing to be answerable for the common good.
In order to overcome today's widespread individualistic mentality, what is required is a concrete commitment to solidarity and charity....
In a society which values personal freedom and autonomy, it is easy to lose sight of our dependence on others as well as the responsibilities that we bear towards them.
It seems to me that the rise of the movement for electoral reform in this country is not entirely unconnected with the culture of individualism the Popes have identified. This is not to impugn the motives of every advocate of voting reform – people who support it do so because they think it is fairer – but one of the factors operating here is the kind of expectations and sense of our own importance that we are encouraged to have in contemporary society. One of the manifestations of our more individualistic and self-centred culture is a strong belief that my vote should make a crucial difference, and there’s something wrong with the system if it doesn’t, so I complain that it is not worth voting, that my vote is wasted, that it is ‘unfair’ and ‘undemocratic’. But it is one of the basic truths of mass democracy that I am just one among millions and in most cases I cannot expect my vote to make a crucial difference. The belief, encouraged by supporters of voting reform, that my vote is ‘wasted’ if it has not swung a close contest, or if my candidate did not win, is a false belief.
From this, one thing seems clear, at least – that contrary to the sometimes outlandish claims of both the Yes and No campaigns, the introduction of AV would be unlikely to make a huge difference to the results of parliamentary elections in the UK, and is similarly unlikely to do anything as drastic as transforming our political culture. It would make only a small difference, but whether that small difference would, in the end, be for good or ill seems to be something of a close call.
It is worth bearing in mind that there is no perfect system – in fact respected economists and philosophers have developed theorems to prove that, if there are three or more candidates, no voting system will be entirely satisfactory.
But we are not looking for perfection, we are looking for the advancement of participation, accountability and the common good, and from this analysis it seems that AV’s chances of delivering these things, while considerably smaller than many Yes campaigners have claimed, are not zero.
In nearly all the areas we have looked at, however, the reasonable prospect of a small advance seems to be balanced out by a genuine risk of damage, in some other way, to those same objectives of participation, accountability and the common good – although not as great a risk as the No campaign would have us think.
All in all, these considerations do not lead us to a very emphatic conclusion on the question of whether the Church’s teaching would encourage us to vote ‘yes’ or ‘no’ to the introduction of the Alternative Vote. The Church’s teaching has given us clear criteria to judge by – the difficulty is that the evidence about how well AV might meet those criteria is extraordinarily mixed.
Peter Scally SJ is Editor of Thinking Faith.
 John Paul II, Centesimus Annus (1991) 47
 John Paul II, Centesimus Annus (1991) 46
 Two examples: ‘It is imperative that no one ... would indulge in a merely individualistic morality. The best way to fulfil one's obligations of justice and love is to contribute to the common good according to one's means and the needs of others, and also to promote and help public and private organisations devoted to bettering the conditions of life.’ Gaudium et spes (Second Vatican Council Pastoral Constitution on the Church in the Modern World), 30.
‘Political power, which is the natural and necessary link for ensuring the cohesion of the social body, must have as its aim the achievement of the common good. While respecting the legitimate liberties of individuals, families and subsidiary groups, it acts in such a way as to create, effectively and for the well-being of all, the conditions required for attaining humanity's true and complete good, including spiritual ends.’ Paul VI, Octogesima adveniens, 46
 E.g. Polly Toynbee, writing in the Guardian, 19 April 2011: ‘Had AV been used for the last 30 years, the only hung parliament would have been last year's – which happened under good old ‘British’ first past the post.’
 Not only do we not know what voters’ second and third preferences would have been, we do not even know what their first preferences would have been, because these might have been different under AV. For instance a Labour supporter in a strongly Conservative seat where Labour usually come third, who voted Lib Dem to ‘keep the Tories out’, would be likely to put Labour first and the Lib Dem second under the AV system.
 Of course, if the Liberal Democrats are particularly unpopular (as polling suggests they are right now) they will not benefit so much from second preferences. If, on the other hand, they are polling well, the question of who Lib Dem voters put as their second preference takes on greater significance, and one might expect that whilst they are part of the current coalition, the Conservatives will receive a greater proportion of their second preferences than Labour.
 In Scotland and Wales, the presence of four main parties makes it even more common for MPs to be elected with quite a small share of the vote. Some commentators have suggested that AV would devastate the SNP’s Westminster representation, since most of their six MPs were elected with 40% or less of the vote, and they – the only pro-independence party – would be unlikely to pick up second or third preference votes from anti-independence voters. In Wales, Plaid Cymru’s three MPs are even more vulnerable in terms of their existing vote share, but less vulnerable on the independence issue, which is less pressing in Wales. Since neither party is perceived as extreme on the left-right scale, however, they could in fact be well-placed to attract the second and third preference votes of other parties.
 It is not surprising, then, that the BNP is supporting the No campaign.
 Survey conducted by the British Election Study at the University of Essex
 Dr David Hugh-Jones, University of Warwick
 It is also possible that in a real AV election, as opposed to a study, many voters will not bother to use their second and subsequent preferences, and simply vote ‘number one’. This is called ‘plumping’ in Australia, which uses AV, and is a widespread practice in those elections in which it is permitted.
 Of course, the Lib Dems would be quick to point out that they are disfavoured by the current system in terms of the number of votes they attract nationally and the representation this wins them in parliament. Even a thirty-seat boost would not be enough to raise their seat total to what they would consider a fair one.
 John Paul II, Centesimus Annus (1991) 46
 The Report of the Independent Commission on the Voting System (October 1998), 126
Cited by the Yes campaign: http://fairervotesedinburgh.wordpress.com/
 See Alan Renwick (University of Reading), The Alternative Vote Briefing Paper, p13
Published by the Political Studies Association,
 Ibid, p7.
 John Paul II, Centesimus Annus (1991) 46
 I have essentially rewritten here a scenario described by David Broomhead, Professor of Applied Mathematics at the University of Manchester, in A formula for fair voting, the Guardian, 22 April 2011.
 John Paul II, Centesimus Annus (1991) 46
 Paul VI, Octogesima adveniens (1971) 23
 John Paul II, Centesimus Annus (1991) 49
 Benedict XVI, Address to the Bishops of the United States at the National Shrine of the Immaculate Conception in Washington, D.C., 16 April 2008
 Arrow’s Impossibility Theorem, see http://en.wikipedia.org/wiki/Arrow's_impossibility_theorem
and the Gibbard – Satterthwaite Theorem, see http://en.wikipedia.org/wiki/Gibbard–Satterthwaite_theorem