3.1 Electoral Systems with Professor David Farrell

Welcome to Real Democracy Now, I'm Nevec Thompson and Real Democracy Now is a podcast for people who think we can and should improve how democracy works. This podcast looks at democracy from different angles to help you think about how democracy might be improved. Thank you for joining me in episode 1 of season 3 of Real Democracy Now, a podcast. Season 3 of the podcast is looking at elections, electoral systems, electoral reform and alternatives. As you can imagine, this is a huge area to cover and I would like to thank Annika Gailia from the University of Sydney who helped me develop a broad structure for this season. I'm going to start season 3, looking at a few areas at a high level before moving into more detail in areas such as electoral systems around the world, negative campaigning and populism, compulsory versus voluntary voting, the various institutions and actors involved and alternatives to elections, where I'll look at sortition and digital democracy. In today's episode I'm talking to Professor David Farrell. David is a professor of politics in the School of Politics and International Relations at the University College Dublin. He is a specialist in the study of parties, elections, electoral systems and members of Parliament. His current research focuses on the role of deliberation in constitutional reform processes. Thanks so much for joining me today David, I really appreciate it. I was looking into electoral systems because it's not my specialty and I came across a definition in the Encyclopedia Britannica online of course, I don't have the book version. He said that electoral systems are the methods and rules of counting votes to determine the outcomes of elections. And I guess to me that seemed like quite a narrow definition and so I just wondered how you define the term electoral system. Yeah, if anything I think that's quite a wide definition because it seems to incorporate the wider panel of electoral laws. How I generally define an electoral system is it's the means by which votes are translated into seats in the process of electing politicians into office. I see, so it doesn't for instance include the age which people are allowed to vote or whether it's online voting or writing on pieces of paper, it doesn't include those sort of things. That's right. I mean, the way we, those of us in this area generally conceive of this is that you have the wider electoral laws and the electoral laws will be when do you get the right to vote whether immigrants or, you know, others and citizens have the right to vote, the way in which the media are allowed to cover elections, how political finance is sorted and so on and so forth. And they're all extremely important, they're all part of the wider what's referred to these days, an issue of electoral integrity. And the electoral system is just one very particular and very important part of that mix. The electoral system is relating to, as I said, the translation of votes into seats. So it's everything from how does a voter complete a ballot paper or, you know, use a computer if it's a computer system, through to the actual counting of the votes and the determination of which politicians or which parties have won the seats. So it's quite a technical definition then, isn't it really like it's not something that the average person in the street would go, "I know what that means." That's right. And that's the challenge for those of us who, you know, work in this area is to try and say to the average person in the street, "Yeah, this is very technical, it's very nerdy, but actually there can be some very important values that underlie the choice of what kind of electoral system you might want to go for." And therefore there is a need to think about it almost in an ideological sense. Hmm. And you've written, I think, two editions already and I understand perhaps there's a third edition. Is it coming out soon of electoral systems, a comparative introduction? Is that right? My publisher thinks it's coming out soon, but I haven't told them yet that I'm still at quite early stages on that. So hopefully sometime late next year it should come out. So how do you actually go about approaching the comparison of so many different electoral systems around the world? It's a massive undertaking and obviously I'm by no means the only one who's been trying to do this. I've been doing it now for several decades. And the way in which generally one does this is you break electoral systems down into the main families of electoral systems and you deal with them in that sense. So you group them into distinct clusters and there are always going to be some weird and wonderful unusual electoral systems that are left out of the mix. But basically the idea is to cover the most significant electoral systems that are used in most of the world's democracies. And so how do you actually characterize those different families of electoral systems? I mean what are the key elements that might make something the same or different? The most important distinction which is the one that has always been used is to distinguish between those electoral systems that have proportional outcomes and those that have non-proportional outcomes. And really that's a range and you can have electoral systems somewhere in between those two extremes. And what do I mean by that? A proportional outcome would mean that if a party gets 30% of the votes then it should get something like 30% of the seats so there should be a degree of proportionality between how many votes a party gets and how many seats it ends up with whereas in a non-proportional system there is often very little relationship between those two things. And so once you start with that basic distinction then you can group the main families along that range between those two extremes. How could you give us some examples of the different approaches? Of the different kind of electoral systems. Well I suppose if we start with the non-proportional ones the most common really, one of the most common still in the world today and the best known is often referred to as first pass the post and it's the system that the British use. The same system is used in Canada and the United States and indeed in India. And the idea there is that you have single seats, districts or constituencies or writings or whatever you might want to call them in whatever country you work in. But basically single seat districts what that means is that every geographical district elects just one politician. And that means that the politician who wins the election only has to get more votes than anybody else, only has to get a plurality, may not even get a majority of the vote. So that can produce some very non-proportional outcomes. We saw that in the recent, in a recent British election where one party got something like four million votes and only one seat in parliament. So that would be probably the most prominent example of one of the main families of electoral systems. And then with the proportional ones there's a lot of different ways of actually delivering proportional representation isn't there? There is and the most common which is used pretty much everywhere in Europe for example is the list system. So it's called proportion or representation list. So list suggests pretty much what the system is about, that the parties produce a list of candidates and in most versions of this list system, the parties will rank order the candidates themselves down the list from number one down to whatever the length of the list might be. And if a party gets, let's say for example, 30% of the votes, it will get 30% of its candidates elected. So that's the system that's used in most European democracies. There's other variants of proportional representation indeed in Australia and Ireland. We use another one which is sometimes called single transferable vote. Sometimes it's called hair clark and that's got a very different kind of logic that underlies it. And I was reading recently about a ballot initiative that was passed in Maine last year to try and change how the voting works there. Are you familiar with that one? Yes, my understanding, I'm not, you know, totally intimate with the details, but my understanding is that they were bringing in a system that you would use to elect your house of representatives in Australia. We here call it the alternative vote, I think you call it preferential voting and the Americans have a name again. It's slightly different from first pass the post because in this case, to get elected the candidate in the single seat district must get an overall majority of 50% plus one of the votes. It's slight improvement on first pass the post, but it's still a non proportional system. So you're saying that if nobody gets more than 50% the other preferences aren't allocated to them all? In the preferential system that Australia uses in the House of Representatives, the idea is that the voters go into the polling station and they rank order the candidates on the ballot paper, one, two, three, four, five and so on. And then they count all the number one votes. And if a candidate gets 50% plus one of the votes on that first count, then that's the end of the count. The candidate is a deemed elected. Oh yes, if they get the majority. You know, 50% plus one, then they eliminate, they exclude the weakest candidate, the candidate with the least votes and they transfer their votes and that continues until eventually somebody emerges with an overall majority. Because of the logic that underlies that system, there's only one person who will get a majority. So that means that all the other candidates in the race in the preferential voting system that you guys use in the House of Representatives, all of those don't get elected and that's why you get a non proportional outcome. So basically you end up, if I go back. to my point about this range between proportional and non-proportional systems, the two most common families of non-proportional systems are first class the post and the preferential voting. And then on the other extreme, the proportional representation systems, they would include the list-electural system used in most European countries, the Herklerk or single transferable vote system used in the Australian Senate and in Ireland, and another system that the Germans pretty much invented, which is nowadays called the mixed-membroproportional, and those are your five main families of electoral systems. I know this sort of maybe goes beyond electoral systems as we've defined them today, but like is the list system more often associated with, you know, multi-party and the non-proportional with like two-party systems, is that, would that be a something that's usual or is it sort of all over the place? No, that's a core question that you're raising and it's an important question. And you know, when I was saying earlier on that there can be some quite almost ideological issues relating to your choice of electoral system, which is why it can be very value-laden as an issue, is that it is known that if you're going for a non-proportional electoral system like preferential voting or first pass proposed, it is known that more normally than not, you're going to end up with something like a two-party system and single-party governments, whereas if you go for more proportional systems like the list systems, you're more likely to get multi-party politics and coalition government. And that produces a very different kind of political system, so it has very important implications for our politics, and that's why those of us working in this area make this point again and again that electoral systems are probably one of the most important decisions that a democracy have to take. And it's interesting, I assumed that Australia as a Commonwealth country and Canada and India would be, would have followed the UK, but obviously we haven't. And I just wonder like, historically, is it a cultural thing or is it the nature of how representative democracy developed in each of the different areas that they've adopted a different electoral system? Because I don't imagine that people, I mean I know once you've got a non-proportional first-passed post, the parties involved want to keep it, because it benefits particularly the two main parties, but was that sort of the motivation in the beginning, or was there some more sort of positive motivation for arguing for first-passed post? There's a certain degree of inheritance that goes into the decision of a new democracy as to which electoral system they're going to adopt, so what I mean by that is there is some patterning to this, the former colonies of certain past great powers are more likely than not to adopt the same system. And so that's why you find Canada, the United States and India using first-passed post because of their British colonial background. And to a degree that also explains the hair-clark system that you guys use because the single transferable vote as I refer to at the hair-clark system as you would say over there is largely a British invention, so it was invented by British men, and one of the unusual features about the evolution of British democracy, almost uniquely so in comparison to the other very old democracies, is that instead of experimenting with new electoral systems on its own people, it tended to experiment with its colonies, and so the single transferable vote or the hair-clark system in parts, one of the strong reasons why we use it in Ireland is because the British decided to trial it on us and we kind of liked it. So when we became independent, we kept it, and in a sort of similar way, there were proponents of new electoral systems in Australia at the time in which you were designing your first political system as a federation who were able to persuade parliamentarians in Australia that the preferential voting system and also the hair-clark system were systems to try, so that's largely why you ended up with those systems. And I think one of the key ideological underpinnings, I guess, as I understand it with electoral systems, is this idea of giving a voice to it to all voters and allowing voters to make a difference, and I've heard arguments that with a list system, all voters can make a difference, I guess, whereas in a first-passed post and even in our system, not so much, if someone gets 51% of the vote, then the other 49% vote doesn't make any difference. And I know in Australia another big thing that we talk about, which I'm not quite sure if whether it fits under electoral systems or sort of maybe electoral laws, is this whole idea of the importance of one vote, one value. But I don't think we take it to the next step to say, well, one vote, one value for 100% of the people, not just 51% of the people. How can people who don't know about this, as you said, you know, we're talking about earlier, you know, you're every day person, evaluate different options, I mean, like what are the criteria that would make something a better system than another? It's, you're raising extremely important and really rather core questions in the study of electoral systems, and the big challenge is because of the technicality and nerdiness of the topic, it really is hard to engage with citizens in discussions about which electoral system would be the better one for your country. And of course, the other feature of this is that because ultimately whatever electoral system you have going to determine which politicians get elected, then politicians are extremely interested in this. And I don't know if you use the adage that we use of Turkey's voting for Christmas, but politicians will be very reluctant to change the electoral system under which they got elected. And so for that reason, they steer well away from any debates about electoral reform. So that's a big challenge and it's very hard to get that right. Now, there have been attempts to try and work around that problem. I think perhaps the most interesting experiment was in Canada in British Columbia in 2003, when the province of British Columbia, the premier set up a citizen's assembly, a random selection, literally a random selection of 160 citizens from the length and breadth of British Columbia, were born together for the best part of a year to debate the electoral system for the province. And it was important to do it that way because those 160 citizens had to become experts in electoral systems before they could adjudicate on what would be the better system for their province. So that's one way in which you can do it. So it's really important that whole area about the rights of citizens under an electoral system. And there are different features to it, which we could of course go into, but I'll just quickly mention one or two of them. One is which was in your question that choice for citizens. So if you are a citizen in a non-proportional system who wants to vote for a green party candidate, you know that the very little likelihood in a non-proportional system that that candidate's going to get elected. So in a sense, your voice is denied. So non-proportional systems deny a certain degree of choice to citizens. That's one dimension. But then there's the second dimension which we might come to, I suppose, which is how much choice does the citizen get on the ballot paper? So in first past the post, you get to just tick then, you know, one candidate's name. You put an x next to the name of one candidate, whereas in the preferential voting, as you will know, you rank order all the candidates. So that means you've got a lot more say over the fate of more than one candidate. And in that sense, the possibility that your vote, full choice will have some input into the final result. Yeah. In the blog post, I was reading about the main initiative to bring in a new system. They were, the person writing it seemed to think that, you know, having this situation where votes of one person got passed on to another, that, and I suppose it's the sort of preference deals that we have here in Australia when there's an election on. They felt that that might lead to a more sort of decent interaction between candidates and less sort of negative campaigning and so on. But I have to say that's not actually my experience of that. I mean, people do preference deals and still bag each other during election campaigns. So I'm not sure that I think there was might have been wishful thinking because they're not actually used to using the system. As what, whilst it's quite nerdy, is there a sort of a slogan bit, you know, like the, the turkey's not voting for Christmas? Is there a slogan or a phrase that sort of explains the benefits of proportional, particularly a list system, which, you know, I find very attractive. But how would you sort of put that to people? Is there a way to explain it simply? And that could be, but it may, it may challenge me to come up with it. But I suppose one way, in which one could look at that is, and there is quite a debate in the political science literature about what makes for a better democracy. And a common argument that is made by a number of political scientists and probably most is that the more consensual your democracy is, the stronger it is. If you have a more consensual democracy, it's more inclusive. So it's trying to bring in all the different interests into the, you know, into the path so that everyone gets a chance to have some voice into decisions in that democracy. And so features of that of as such a democracy, a consensual democracy would be coalition governments which are run properly. So they would be, you know, strong coalition governments and, you know, not too many parties, but at least some parties perhaps. A parliament that is listening to the opposition not always excluding the voice of the opposition, so strong committees in your parliament. And other features, but in the heart of all of this would be in proportion. representation system. So that's why most of us who work in this area, I can't say everyone, but most of us would say that if you held a gun to our heads and said what would be the best electoral system to go for, we wouldn't be able to answer precisely what electoral system is the best because there is no such thing as an ideal electoral system. But what we can say is that proportional representation is certainly better than non-proportional representation. Are there electoral reforms that you think warrant serious consideration that haven't really been used anywhere that are sort of thought experiments at the moment? I don't think there are. I mean there are features of existing electoral systems that are quite attractive and it's sort of trying to pick out the best features of those electoral systems and some people refer to a sort of sweet spot. So sort of where roughly would you want your electoral system to lie that would make it as good as you can get? And features of the sweet spot would be you wanted a system that's proportional but not too proportional. So you want districts that are electing maybe five politicians and no more than that so that there's not this sense of cutoff between the politician and the and the citizens. You want an electoral system that gives choice to voters so on the ballot paper. So that would suggest to me you don't want closed list where the voters can only vote for parties. You want a system much like Hare Clark where voters can rank order the candidates in the parties and therefore if there's a particular if a party has a list of candidates and I really like that candidate I can put them as my as my number one. So there would be certain features like that that would lend themselves to coming up with a good electoral system but honestly I can't think of any particular feature of an electoral system that hasn't been designed already. That would be ideal to include. I can see what you mean about sort of the areas being too big because as you would know the Senate in Australia is on a state basis which in Tasmania might be quite okay but in New South Wales it means that I don't think anybody really feels particularly attached to their senator you know on the basis that it's a New South Wales senator it's almost too big which leads me to another question that I'd like to throw in. There's a lot of talk particularly here in Australia for various reasons but I know around the world about the idea of using sortition to fill parliament and in particular some people talk about doing that for the House of Review so in Australia it would be for our Senate. Do you have a view on you know that as a sort of electoral reform I guess it doesn't relate to it's not an electoral system reform but it's certainly an electoral reform. It's certainly a debate that is beginning to get some traction. It's a very interesting idea and ultimately this idea of sortition or random selection is what underlies the growing interest in citizens assemblies or there's sometimes called mini-publics and actually here in Ireland we have these things been experimented right now so we have an Irish citizen assembly that was established late last year by the Irish government of 100 citizens selected at random from the population by a market research company and they're spending a year discussing a number of very important constitutional reform questions for our country so it's a method that has some merit and I have allowed sympathy for the the method. Whether I go so far as to propose that you replace your House of Review your second chamber with a sortitionist chamber I'm not sold on that idea. I think sortition has a place but I'm not certain if it if it's places to be a second House of Parliament I think that might be pushing it a bit. I understand I think it's in Mexico or might be another Latin American country where parties have both electorate people so they've got people politicians are elected to represent an electorate and then they have list members that are elected based on the proportional the number that the party gets and there's one particular party that has chosen to actually fill its list positions through random selection amongst its membership have you heard of that? I haven't heard of that particular example I mean that electoral system you're talking about sounds like a version of the German mixed member proportional system so tries to mix first pass the post with a list system and that's become variants of that have become quite commonly used now in a number of countries where you're going there is really into the question of how candidates are selected by parties and so political parties are you know facing their own particular challenges and so different experiments are being tried with how they might open up the candidate selection process and some other way you know the traditional classic way was the party apparatus the you know the party elite and the party activists decided who the candidates were going to be now we're moving into a world where parties are moving towards something like the American primary system to select their candidates and something like the experiments you're talking about in that Mexican case would be an interesting variant of that and I know you've said to me that there's no such thing as an ideal system although you did talk about a sweet spot and so I wonder you know if you were asked to redesign the Irish electoral system what would it look like would it be pretty much the way it is now or would you be looking to try and change it to get that sweet spot yeah I mean I think the Irish electoral system is close to to to good and indeed the the article there is a there's an academic article published by some colleagues who are not Irish a few years ago about the sweet spot and they they make that point that Ireland's single transferable vote system is is roughly in the right place as it happens the Irish government two years ago another Irish government set up a constitutional convention so a version of a citizen's assembly this was a random selection of citizens sitting down with parliamentarians and one of the topics that they discussed was the Irish electoral system and that convention made some recommendations about how to tweak how to change features of the Irish electoral system which I think had some merit if such as making our districts bigger or you refer to them as electorates but I present our our parliamentarians are elected roughly three or four in a constituency as the average three or four in an electorate and I mentioned earlier on that the sweet spot argument would suggest you want to be somewhere around the area of five parliamentarians so one recommendation of the constitutional convention was to increase our district sizes our electorate sizes to five and the other was to move away from a ballot paper design that's very much based on our British colonial inheritance which is the alphabetical ranking of the candidates so when we go into the polling station our ballot paper is the candidates are listed alphabetically and the recommendation of the constitution convention which to explore other ways such as the system I believe you use in the ACT where you rotate the ordering of the candidates and there are other variants again and things like that could be done to to fix some some problems some issues with our system. Yes in most of the elections in Australia it's a draw it's a random draw as to what order you are on the ballot paper and interestingly for my research my PhD research I was interviewing some councillors in Queensland and and they'd actually recently had their local government area changed so that they didn't have wards the local government area was like a list system and so it was sort of the top I think there was six the six councillors and the top six got selected and the previous mayor was saying to me that he felt was almost like sortition because people might have one person that they were interested in and they'd give them their one and then they'd just go down from the top and allocate the numbers just two three four five whatever and so he said you know it was really the luck of the draw as to where you were on the ballot papers to whether you got elected or not it wasn't actually I thought that was quite an interesting take on the positions on the ballot paper and that impact of the list model yeah and I mean I don't know of that level in Australian elections but I know more generally in Australia you you tend to have a situation where voters are required to rank order all the candidates and that's unusual you know that that has its drawbacks and one of those drawbacks is that voters will I don't know sometimes it's referred to as donking you know pinning the tail on the donkey type voting that voters just don't care they want to express the vote for the candidates they want but they know their ballot paper will be declared invalid unless they rank everyone so they just randomly select in our in our version of that system voters only have to express one preference they don't have to express them all they can express as many as they like but they only have to have one preference so yeah no we do we used to have to number them so for instance in our senate which is like this huge ballot paper if you wanted to vote under the line you had to number every single one in you know your order of preference but I think in the last election they changed it so you only had to do the number that that would could be elected so that you know it was a little bit simpler but it's still yet certainly put people off actually expressing their opinion about individuals yes thank you for joining me today in next week's episode I'll be talking to Emirates professor aren't liepard about his lifetime's work I did research in the 1960s on my native country the Netherlands and I was struck by the fact that the Netherlands which is a or at least at that time was very much a divided country divided terms of religion especially for Protestant versus Catholic versus non-believers and in terms of political ideologies as well. And the theories about democracy at that time were saying that democracy is very difficult to achieve or it's very difficult to work well when you have deep divisions like that. And just looking at the Netherlands, I would say, it was at that time a very well governed stable country, so both the democracy that has happened in an environment that according to the theories and political science at that time basically could not exist. And so when I started looking at that question more carefully, the explanation that I came up with is that in such deeply divided societies you can have democracy work well if there is a hearing of power, making decisions together instead of having one group make the decisions and the other group opposing it therefore leading to conflict. So that was step one, then I looked at other countries and found that pretty much the same pattern could be found in a number of European countries, Belgium, Switzerland, Austria and also in some non-European countries like Lebanon and Malaysia. So it then became a general theory of how democracy can work in divided society. One next step was to think how can this kind of democracy which I called to use a technical term, Consulciation of Democracy or to use less technical term, power sharing democracy, and we used that in order to deal with democracies generally. And I started making this distinction between majority in democracy and what I then feel called Consulciation of Democracy. But I changed that terminology a little bit later on to make it more general and so the contrast then was between between mature theory and democracy and consensus democracies. And that became the basis of the work in my book called Democracy and then also in my later book called Patterns of Democracy. I hope you join me then. Thanks for listening to Real Democracy Now. You can find more about today's topic in the show notes at www.realdemocracynow.com.au. If you enjoyed this program please subscribe to this podcast, write a review, share the podcast with your friends and join the conversation on the web page or on Facebook or Twitter. I'd love to know what you think is the essence of a real democracy. Real Democracy Now, Real Democracy Now.