← New Books Network
New Books Network
Georgia Kernell, "Inside Parties: How Party Rules Shape Membership and Responsiveness" (Cambridge UP, 2025)
00:49:48
Loading summary
Transcript37 lines
- [00:01]
A
Welcome to the New Books Network.
- [00:05]
B
Hello and welcome to the New Books Network. I'm your host, Satiki Barua. Why are some political parties more successful than others at recruiting members and developing policy commitments? This is the central question at heart of Inside Parties, a new book by Georgia Cannell. While political scientists have long studied party organizations and electoral competition, we know surprisingly little about what happens inside political parties themselves. The book addresses this gap by examining how internal party rules and procedures shape party membership and political responsiveness. Drawing on an original data set of party statutes covering 65 political parties across 20 parliamentary democracies, the book offers one of the most comprehensive studies of party organizations to date. Georgia Cannell is an associate professor at the Department of Communication and Department of Political Science at ucla. She teaches courses on political communication, media and politics, and research methodology. She is also a research affiliate with the Future of Democracy Initiative at the UC Institute on Global Conflict and Cooperation, a faculty affiliate at UCLA center for European and Russian Studies, and and the director of UCLA Political Party Organization's Lab, something we'll come back to later in our conversation to talk about our book. I'm delighted to welcome Georgia Cannell to the New Books Network. Georgia, congratulations on the book, and thank you so much for joining me today.
- [01:40]
A
Thank you, and I'm happy to be here.
- [01:42]
B
I would like to begin by asking about the origins of the book. You open with the debates surrounding Brexit and argue that internal workings of both the Conservative and Labour parties played an important role in shaping those events. You suggest that internal rules governing these political parties can have a profound consequence for their electoral fortunes. So can you take us back to the moment that sparked this project? What led you to look inside political parties to venture as you put, into the secret garden of party organization? And what convinced you that these internal roots deserve much greater scholarly attention?
- [02:26]
A
Yeah, so my interest in parties, internal organizations stems back way before Brexit, at least this most recent Brexit successful vote. I was an undergraduate at UC Berkeley and I had taken some courses on I had taken a class on political parties with Henry Brady, and I really loved the class. And I had realized that I really liked studying countries outside of the U.S. i went to undergrad not knowing much about how things worked outside of the US and the more I took on comparative politics, the more interested I was. And that parties course, I think we compared party systems in five different countries, and I found it to be so exciting to see that parties don't look the same in countries outside of the US and that still is the case Today they have different histories, they represent different groups of people. There's a different number of the parties, and they work differently internally as well. And so as an undergrad, I had also taken a methods course that had introduced me to qualitative and quantitative methods and causal inference and thinking about what that means. And I had a pretty good idea. I wanted to go to graduate school and study comparative politics and perhaps parties. So I approached this methods professor, and I asked him if I could help him out as a research assistant. And he said no, he didn't need me. But he was very kind about it and suggested that I contact or that he put me in touch with one of his graduate students who was getting their PhD studying political parties. And so he put me in touch with somebody named Ken Green, who's now a professor at UT Austin, but at the time was a PhD student. And Ken invited me to be his first research assistant as an undergraduate. And Ken was studying Mexico. And I was very interested in studying Latin America because I spoke Spanish. I thought I spoke Spanish fairly well. I'll get to that in a minute. And I started working for Ken. And what he was interested in was studying how the pre. Which was the dominant party in Mexico, but was losing its. Its foothold, how it. It's how it was organized internally, and then also some of the rules in the. The two competitive parties, the PUN and the prd. And so, I mean, his general focus actually wasn't on an internal organization, but my job was to collect the statutes and to read them and to answer some questions for him about them. Now, in doing so, I learned that I don't speak Spanish as well as I thought. It was very difficult to read the. The party rules in, in a different language and understand them. And the language skills I had gathered didn't prepare me for it. And there was no such thing as Google Translate back then, so I had to actually use a Spanish English physical dictionary to go through these statutes and understand how they worked. The process was really exciting, though, and I. That's what got me originally interested in political parties. And also just how. How much variation there was within a single country was quite exciting. Yeah. So I went to graduate school to study parties in Latin America. And I thought maybe Mexico would be my main case because I had learned about it and I loved the work that we had done or that I had helped him with, which probably wasn't that much. I don't know if I helped him that much, but I loved what I did. And. And so I went to graduate school And I, I started by taking classes, as everybody does, and I worked with my advisor on a cross national project that was not really about, it was about partisanship, it wasn't about party organization, but it was looking at lots of different countries and that was very exciting too. And so then when it came time to write up my dissertation proposal, my advisor gave me a list of 20 books to read that had to do with party organization. And I read them, we came back, we met and I discovered in talking with him and reading those books that nobody had really asked the questions that I was interested in anywhere, not, not in Latin America, but also not anywhere else, certainly not in countries outside of the US and those questions were about how do, just basically first and foremost, how do parties work? Who selects the candidates, who selects the leaders? Who can remove a leader? How are the platforms written? How are resources allocated? These basic things about how any organization functions. It could be parties, it could be a church, it could be a school, any level of government. They really weren't cataloged in a systematic way yet, even in other countries. And for students of comparative politics, we all know that much of our research on democracies is, starts in Europe. And so I kind of assumed like this had been done in Europe and that I was going to be studying it in Latin America. It turns out it hadn't really. Yeah. So that's why I decided to start studying Europe and parliamentary systems instead of Latin America. I thought it would be like a first step. And then I, I never went back. So now I'm a Europeanist.
- [07:53]
B
One thing you mentioned is that there really wasn't a systematic understanding of how party organizations function internally. We know a great deal about elections, party competition and voting behavior, but much less about the internal organizational mechanism that shape how parties actually operate. I'm curious to know that the gap in the literature, how that gap in the literature shaped your own research agenda, in what ways did recognizing that absence help you chart the direction of this book?
- [08:27]
A
Yeah, I mean, I'm not the first to study internal party organization, but I'm the first to study it in such, in the way, in this systematic way, especially of looking at these particular rules, which I'll get into, but of candidate selection, leadership selection and manifesto design. And I think it was absent because it requires a lot of work, especially before these statutes and rules were digitized. So when I started collecting these data in the early aughts, the majority of parties statutes were not available online and some of them were not available electronically at all. So I decided to study Parliamentary systems. I limited it to parliamentary systems because most established long term democracies had parliamentary systems, but also just because they didn't have the time and resources to extend to what are called semi, semi presidential systems that have both a president and a parliament, an independently elected president and a prime minister elected by the parliament or to presidential systems. And so when I started, I contacted all of the parties that I was interested in. I created a cutoff of 8% to look at parties that had a very good chance of making it into government or that had made it into the government. I looked at parties that had 8% of the vote in the last parliamentary election and then I contacted all of those and that turned out to be 65 parties. That's where that comes from. 65 parties in 20 countries. I contacted them by email. If I couldn't find their statutes online, which was the case in the vast majority. I contacted them by phone and in some cases I just went to the party headquarters if I couldn't get a hold of them. And it was difficult. I mean, some places would mail me statutes. In Ireland, I think two out of the three parties sent me statutes in the mail and I had to agree to not share them. Now they're available, publicly available, and even the historical ones are. And so I, I was able to, you know, conduct that research and publish what I found about those parties. But they were these nice, very beautiful statutes that were clearly meant for being handed to, to be handed out only to members of the party and not made publicly available. And that was the case in a few of the other various other countries as well. So I think it's just, it was difficult to collect. But the people who have collected the data before, there are people who have collected huge cross national data sets and on parties, but some of them, they either focus on different aspects of the parties. Ken Janda, I'm thinking in particular collected a huge data set on different aspects of the parties or, or didn't look as, as much over time. So I look over 20 years of the parties, organizations and previous studies like Katz and Mayor's study, look at a particular, a shorter moment in time and fewer, fewer parties or fewer countries, I should say.
- [11:41]
B
So One of the things that immediately stand out in the book is the incredible amount of work that went into compiling this data set of party statutes. That is an enormous contribution in itself. Building on that, you develop a theoretical model linking internal party rules to organizational outcome. So let us begin with candidate selection. Why did you see candidate selection as an important function of the party and how does your model reveal about why these internal rules matter so much for how parties function? And then we can move on to membership, which is another central part of your argument.
- [12:25]
A
So I started with some preliminary interviews in Portugal and Spain. My Spanish was good enough, I guess, to, to start with those interviews. And then I had a great translator in Portugal and I started by just asking a lot of people in those countries from the, the largest two parties in each of those countries what they thought were the most crucial rules that governed how the party worked, that determined the allocation of power that affected which coalitions they would join, if they would, that would make them more or less competitive, etc. And the people I interviewed in person in Portugal and Spain, as well as on the phone and other locations continually pointed to the rules that affected candidate SEL and so by this I'm in particular I'm particularly talking about parliamentary candidates. So I'm looking at parliamentary systems here where members of the lower house are elected to parliament and then they can select the prime minister. And in some of these cases we have, we typically see single party majority governments, and in other cases we rarely see that and see either minority party governments or minority or majority coalition governments. And so candidates, candidate selection is the process by which a party chooses which people will make it onto the ballot. If you have an open list system, proportional representation system, then what that means is that there's a list of people under the party label. And when people, when voters go to cast their ballot, they select a person from their district and then seats are allocated first to parties proportionally and then to individuals, individual candidates in the descending order of the votes that they received. So even in an open list system like that, even if you're voting for a person, you're first and foremost voting for a party. And then the candidates are elected and after that, as long as the, you know, in proportion to the share of the seats that the party received. In a closed list system, people just vote for the party and the candidates that make it in are written in, in order by the party and the party somehow chooses the list of which candidates will make it in, which ones you know, are top of the list, which ones are the lower point on the list. In single transferable vote systems, there's also a list of candidates, but people can rank them. The same is the case in alternative vote. Except for an stv, there's multiple candidates that can be chosen. In av, there's only one. And in single member plurality districts like in the UK or the US or Canada, there's just a single candidate that people choose. There's also mixed member systems where there's a mix of both PR and single member districts. And so all of these are electoral rules that differ across countries and that have been studied in great depth, that have been shown to really affect things that we care about, like representation and responsiveness. But the same thing is happening within parties as well. So parties are selecting somehow who gets onto that ballot in the first place, who is on that list of people that then voters choose among on their ballot for an open list seat, or what the who is on the list and what the order is in a closed list system, or who is the sole candidate that represents the party in a single member district, and so on. And that process by which parties are selecting the people to their to to make it onto the ballot. And in some cases the order on the ballot is what is called candidate selection. And it's what we don't know a lot about or we didn't know a lot about. And so when I interviewed people, they talked a lot about candidate selection, how candidates get on the ballot, the rules that affect that, and in particular, the tension between giving regular party members control versus leaders control. And as I continued to do research and exploratory interviews, I realized that this tension between the members and the leaders seemed to occur in every single party and be at the forefront of people's minds when they thought about institutional change within parties.
- [16:52]
B
So something I found particularly interesting is that the tension between party leaders and party members doesn't just appear in candidate selection alone, but it really runs throughout your discussion of membership as well. So could you talk us through that relationship? How does the balance of power between leaders and members shape the way parties are organized? And what does your model tell us more about the role of membership that plays in shaping party policy?
- [17:23]
A
EFFY so this tension between members and leaders I decided to really investigate. And I found that in some parties, a lot of control is given to members. So party members are people who like the party, they vote for the party, and they go beyond just voting for the party. They pay a membership fee. Typically there's a few cases where it's free, but they pay a membership fee and they sign up to be a member. And that signing up process typically also involves them saying they won't be a member of another party and that they pledge some sort of allegiance to this party, at least that they're not going to join another party at the same time. And they pay this fee annually or monthly, and it gives them access to the party. What that access looks like differs significantly across parties. In some cases, like I said, parties give these members a lot of control. In fact, the candidate selection process in some cases is completely up to members. So joining a party gives somebody not just camaraderie, an avenue to meet other people, but also a way to affect the party's position by selecting candidates. In other parties. There's no powers that are given to members like that, but instead, the party focuses really on this camaraderie or these social elements. So parties might have special meet and greets for members where they could meet candidates or elected officials. They are invited to attend rallies, they're asked to hold fundraising and get out the vote drives, and they're kept in the party loop. A lot of people join parties because of this information advantage and the social aspects that come with joining a party. And if people really like a political party, they want to show that they like it. So they want to be a member. They want to have that badge, that bumper sticker, that sign that they're a member of the party. And so this difference in the, in the role that members play, I hypothesize, has an effect on how candidates are, not just how they're selected, but on who they actually are and what they represent, and therefore what the party in general represents. And I argue this is the case not just for candidate selection for parliament, but also for selecting the leader of the party, as well as the design of the party manifesto, which is its platform or statement of policies that it plans to pursue if it's elected and in government. And so I look at those three factors in particular, candidate selection, leadership selection, and manifesto design. And I look at the level to which party members, as opposed to leaders, have actual control over those three areas in the party. And I argue that in parties where members don't have this control, where they don't select the leader themselves, where they don't get to vote on policies that make it into the manifesto or choose the order of candidates on the list. And those parties, which I call centralized parties, because they're run by the central level of the party, the leadership people will, people who are, would be members. That is, voters will join the party if they really like the party, but they won't join it if, if they're just voting for it because it's the best option out there. And that is, the more the, the, the closer a person is to the party on an ideological dimension, or the closer that they feel to a party, the more likely they'll join it because they really like it. They want that camaraderie, like I said, they want to show their support. And that's really what will just solely motivate somebody to join. So that's not surprising, that's intuitive. We would expect people who like the party to be more likely to join it. But I'm arguing that that's really the only thing that matters for people whose the party they prefer is very centralized when it's controlled by leaders. In contrast, if we think about people who support parties that do give them opportunities to vote for the candidate or the leader or the manifesto, now there's two motivations that might lead somebody to join. They might join because they really like the party, same thing as in the other case. But they also might join because they want to change or affect the party's policies or candidates or leadership. So a person now who maybe doesn't agree that much with the party, but still it's their favorite party of the group or the party they're going to vote for, they now might have an incentive to join, to try to select candidates that are more in line with their own preferences. And so a person who's more liberal than the party that they vote for, significantly more liberal, let's say, will be more likely to join that party. I argue if that party is decentralized than if it's centralized. If it's centralized, then joining is just really something to, to show that they're joining or something for personal benefits. But it doesn't give them this, what I call instrumental benefit of changing the party. And so what this means is that the type of parties, the, the type of organization that a party has that is decentralized or centralized will not only affect sort of how much power it gives to members, but it will then affect who those members are in the first place. And the hypotheses are born out of a game theoretic model in the, in the book, in, in a chapter in the book. And so I predict that parties with decentralized candidate selection will attract more extreme members than parties with centralized candidate selection. And the same should be true for leadership selection and manifesto design. And not only will their members be more extreme, but because those members have a say, the parties themselves will be more extreme. And so by giving power to members, parties allow their candidates and their platforms to be pulled to more extreme positions than had they not given them power. And that sets up this paradox. Well, why would parties ever give power in the first place? Wouldn't leaders want to control everything? Well, there's a reason that they give power, power to members. One might be ideological. We see that green parties definitely give power to their members because it's actually built into their ideology, that is that everything should be grassroots and controlled from the bottom up. And but other parties also grant power to members, and often they do so to increase their membership, to increase people who will support the party, who give money to the party, and who will be there to advocate for the party during elections. And so there's this tension for leaders between wanting to grant more power to members so that they can increase their membership, especially because membership is really dying off in today's day and age. But at the same time, they may want to constrain the power that they give members, because giving power to members can actually change the party itself, the positions that it advocates for, the candidates that it selects, and the leader who runs the party.
- [24:35]
B
So your model makes two really striking predictions. On one hand, you say that decentralized parties are more likely to attract members, but those members also tend to come from the ideological extremes. And on the other hand, centralized parties appear to be more responsive to the broader preferences of voters, although they don't allow power in the hands of the members themselves. So what explains these differences? What are these organizational mechanisms that lead to such contrasting outcomes between decentralized and decentralized parties?
- [25:18]
A
Yeah, well, so what you stated is exactly right. As far as those are my hypotheses, right, that I expect decentralized parties to have a more sizable membership, but also a more extreme membership and that that can then pose a trade off to parties because that extreme membership can lead to more extreme positions held by the parties. And this is really easy to think about in a two party system because extreme means going more to the left for the left party, more to the right for the right party. But in a multi party system, it's not as clear what extreme means for a party that's in between, located in between other parties, like on a left right spectrum. And so I won't go into too much detail here, but when I say extreme, I mean extreme relative to the position of the government. So any party that's still left of the government, its members are going to be more to the left of that party than they are to the right. All right, so those are the hypotheses. And then I test them using cross national data on party organization and survey data, as well as through an experiment. So I collected this data on how parties are organized using party statutes and their constitutions. As I mentioned earlier, I collected Those sometimes hard copies of them. But in the more recent time period, I was able to access most parties rules online. The majority of parties make them accessible now or they're pretty easy to obtain. And so I collected statutes and constitutions and bylaws from parties over a 20 plus year period for all of these 65 parties. And I then cataloged the level to which they're decentralized or centralized on these three dimensions of candidate selection, leadership selection and manifesto design. I also coded parties based on whether or not they grant members a vote, even if that vote is only advisory or in practice it doesn't really take an effect. So in some cases parties will give members a vote, but it doesn't carry any weight. The party themselves will select the leader, for example, or in other cases they give members a vote at the final stage. And it's often an approval vote where there's only one candidate. And so I also look to see whether voting itself matters or if it's really having a say. And then I compare or I pair this data on party organization with cross national survey data from three main sources, the European Social Survey, the International Social Survey Program and the Comparative Study of Electoral Systems. And I and I look at whether or not parties that are more decentralized have more members and the positions and if those positions of those members are more extreme, and I find strong support for the hypotheses. I also look within parties at several member surveys to see are indeed people thinking about these instrumental benefits when they're joining. And I find that they are. A lot of people state that they join decentralized parties to make a difference or quote, to change the party. And then I ran a survey experiment of people who are long term voters for the Labor Party but are not currently party members. And I told them that the Labor Party was going to either change its rules to be more decentralized and give party members a greater say, or to be more centralized and give them less of a say. And then I compared that with those who I didn't say anything about the change in rules. And I found that hearing that the change in rules was going to be more decentralized led to an increase in membership. But it did so particularly among people who were to the left of the Labor Party. So those people who are more extreme were more likely to join the party when they knew it was decentralized and they were less likely to join the party when they heard it was going to be centralized or to say that they would join the party. And that further supports this Argument that decentralization increases membership because it gives people more power, so there's more bang for their buck if they're joining the party. But it does so disproportionately among those who hold extreme positions. Hey, I just Venmo'd you for rent.
- [29:50]
B
Nice. Now I can instantly spend it, whether
- [29:52]
A
I'm checking out online with Venmo or
- [29:54]
B
using a Venmo debit card.
- [29:56]
A
Say more. More.
- [29:57]
B
Exactly. Because the more you do with Venmo,
- [29:59]
A
the more you get like earning up to 5% cash back with Venmo Stash on a bundle of brands. So order more pizza.
- [30:06]
B
The math demands it.
- [30:08]
A
Get the Venmo debit card.
- [30:09]
B
Venmo Stash bundle Terms and exclusions apply. See termsnmo me dashterm venmo checkout not
- [30:14]
A
available at all merchants venmo mastercard is issued by the Bancorp Bank N.A.
- [30:19]
B
another aspect of the book that I really appreciated was its methodological design beyond the data set itself. Of course you're combining large scale quantitative analysis with qualitative evidence interviews and experiments. But what really intrigued me was this multi method approach. So why was it important to bring all these methods together? In particular, what did experiments allow you to demonstrate that observational evidence alone couldn't? And how did combining these methods, different approaches, strengthen the causal explanation behind your model?
- [30:57]
A
Yes. So I derive the hypotheses from a formal model of membership and then I test those using observational data. I should mention. I also look at the World Value Survey here and the observational data can suggest a relationship, but they can't show us that there's a causal effect of party rules on membership. So the, the results are supportive of the causal argument. But there could be something else at play. So the reason why I run that experiment that I mentioned with the Labor Party is to to manipulate what it is that that otherwise similar labor voters hear. And so that if I see a difference between those that are between the treatments, whether or not people are told the party is centralized or decentralizing, if I see a difference in their likelihood of joining, then I can know that it's that treatment that's causing that difference instead of something else, some other attribute about those individuals. And then I test these long run implications of party responsiveness by then using computer simulations to see if decentralized parties are indeed closer or farther from the vote maximizing position that they could adopt. And so the methods include the mathematical model, the observational data, the experimental design, and then these computer simulations. But I have to say, probably the most Interesting method for myself in this book was the qualitative work of the surveys that I or of the interviews that I conducted. So I met with over a hundred different party officials. These were typically the general secretary of the party, that is, the person who's really in charge of running the party on the ground, the mechanics of the party. And I conducted pretty long extensive interviews with these individuals and learned just a ton about how they think about party rules, how others in the party think about them, what the tensions are within the party. And I found that to be, for me, the most exciting and illuminating part of the book.
- [33:10]
B
Yes. So I think that's one of the real strengths of the book because it demonstrates how different methods can complement one another. And I imagine that it will encourage many party scholars to think more seriously about mixed methods research. Something else that struck me while reading the book is that there doesn't seem to be an ideal organizational structure. So while every institutional arrangement appears to involve, let's say, certain trade offs, and we spoke about how decentralization encourages participation, empowers members, but it also risks attracting activists from ideological extremes. And centralized parties seem to be better at responding voter preferences, but it of course concentrates power among the party elite and can alienate members. For instance, you say that, you point out that younger members are more likely to be alienated in these centralized parties. So how should we think about these trade offs from the perspective of democracy? What do they tell us about representation, inclusiveness, and the broader role of political parties in democratic systems?
- [34:21]
A
Yeah, the book really does have a fundamental argument about internal democracy, and that is that parties that are internally democratic are not necessarily more representative of the general population. And so having internal democracy may not be a good thing, both for the parties in getting elected and for voters in being well represented. If we think that we want parties to represent everyone, and not just those that are the most interested and the most likely to join and participate and put their time into the parties. Then having a centralized party where leaders select candidates, write the platform, whether it's behind closed doors and smoke filled rooms or not, might actually be more representative and better for democracy overall. And this is something we've seen before in research, not about political party organizations, but in other areas, especially about direct democracy alternatives. So I can't say that any direct democracy option is problematic, but when there's direct democracy options, that, and I say options, meaning when people are allowed to participate but have to self select into that participation, for example, to vote on propositions or a referenda within states, we know that only a subset of people participate. And so any type of democratic system that gives more power, but only to those who cross that extra hurdle of participation, may in fact undermine democracy overall. And the message for parties is that internal democracy is not always going to be beneficial. And actually I have to say that party, that message does not fall on deaf ears. It's the mem, it's the party bureaucrats and leaders that told that message to me more than vice versa. I can give a quote from a candidate in Sorry, the party president from the Australian Labor Party here. This is from an interview in 2016. And he says a risk in giving members more power, particularly to determine the party leader, is that you end up with far more left wing, in our case right wing. Presumably for Conservative parties leaders who then present an image to the electorate that is not attractive. Moderate and Conservative labor leaders are pointing north to Britain as an example of how all of this can go wrong. And that goes back to the start of the book where I compare the Labour Party in the UK with the Conservative Party in the uk, The Conservative Party has rules that allow the members of Parliament to quickly replace their party leader. And so we saw alternations in power among party leaders in the Conservative Party at times when the leader would become unpopular. And these were very quick to happen. And as a result the Conservative Party itself didn't lose popularity when its leadership did. I contrast that with what happened in the Labour Party over that same period of time following Brexit, where Jeremy Corbyn, the leader, refused to stand down even as many of his shadow cabinet members and previous supporters resigned from their positions and said that they no longer supported him. In fact, a majority of members of Parliament in the Labour Party stood against the labor leader and yet he won even more support among party members in the next leadership election. And that's because the Labour Party gave power to members, not to members of Parliament, that is to people who were interested in joining the party, paying a fee to do so, but not those who are actually in office and who presumably represent a larger share of the population. Now we've more recently seen that the Labor Party has been able to deal with Keir Starmer's dropping popularity ratings. But how they did so. But they did so because he self selected into removing himself as the leader. If he hadn't, however, we probably would have seen the same thing play out for the Labour Party leadership where an unpopular leader drags down the party again.
- [38:45]
B
So one of the fascinating implications of your findings and what you just mentioned is that it almost seems like a paradox. So parties that are more internally democratic necessarily end up being more representative or responsive to the electorate as a whole. And that's such an interesting finding. But that also raises questions of how far this framework travels. Your evidence comes primarily from Europe, but do you think that these organizational dynamics extend beyond that context? How applicable do you think your framework is to regions like Asia and other parts of the world with very different party organizations?
- [39:30]
A
I think the general tension is there in all cases outside of parliamentary systems where the more power that's given to party members, the more extreme members might be attracted to join the party in the first place. And so we see this, actually, I'll take the case of the United States. We see this discussed a lot with respect to primary elections. So primary elections are basically very decentralized forms of candidate selection where anybody who doesn't even have to pay a fee can participate in selecting the candidate from their party's list of potential candidates to then appear on the ballot come November. And these primary elections attract a set of voters that are not representative of the general public and that tend to be more extreme. And so criticisms of the primary system in the US as well as primaries outside of the US in other countries, are that they drive left party's candidates to be more liberal and right party's candidates to be more conservative. And so we see a competition in the general election of candidates that don't represent the median member of society or even the median member of the party, but rather median or representative candidates of the most extreme groups within the party. And so this should play out truly in any case, where we, where parties are selecting candidates to then compete, when they give that power to interested members of the public, then we should expect those candidates to be more in line with their interests than the general electorate. That being said, there are some ways in which primaries may be more or less accessible to that general member, to those general members of the electorate. And so the more accessible the primary, the more likely they are to align with other elections and the greater the turnout. We should expect this to have a weaker effect.
- [41:35]
B
So basically what we are seeing is that parties that are, let's say, more democratic in terms of political participation are much less representative and responsive to their electorates in comparison to parties that might not give their members such an avenue for democratic participation. Right. So this is an interesting finding. And as we already discussed and yeah, that's exactly right.
- [42:02]
A
And it's the ones that are not just perform a democratic or that look like they're Democratic because They give members some vote. It's the ones that really give the power to members, where members actually control who is the candidate leader or what is written in the manifesto, not those that have some sort of symbolic power. But yeah, we see that internal democracy undermines representation.
- [42:30]
B
Yes, and that's such an interesting point because it really challenges our assumptions about internal democracy. Before we wrap up, I'd like to return briefly to the methodology. I know we already spoke about it, but one thing that of course really interested me was how successful you integrate these various approaches, large scale quantitative analysis with experiments, interviews, into this one single project. So from your own experience, why was this method, this multi method, necessary? I'm asking this because I think many scholars working on political parties are increasingly looking for different avenues and methods of research. And I think your experience would offer a valuable example of how these approaches can complement one another.
- [43:15]
A
Emphasize again the importance of the interviews in the book, both for a reader, but also for myself. In writing the book, when I started out, I thought I was going to conduct interviews to sort of add some flavor and examples to an argument that I was coming up with myself. And in fact, the interviews changed everything. As I learned about how parties were organized, I decided because of that that I wanted to model membership the way I did. And I created the formal model in response to the and because of what I was hearing in these interviews. So I can't wait to get out there and conduct more interviews. I find them to be really thought provoking and helpful with hypothesis generation, not just ways of providing examples of hypotheses that I've already come up with and tested. And that was a surprise.
- [44:13]
B
That's a really valuable insight and I think it certainly helps. It will certainly help researchers who are considering adopting similar approaches. So before we finish, I would also love to ask about the Political Party Organizations Lab which you're currently directing. Could you tell us more about what you're building there and how do you see it shaping not only your future research, but the broader study of political parties or the internal rules over the years?
- [44:44]
A
Yeah, so like I said, when I started, I had no Google Translate and no AI. So I needed the help of native speakers to read these statutes, translate them and code them. And gosh, I forget, I think there were 10 different languages that I needed to hire people who were native speakers and readers of from. And so there was just a ton of work that was done by people other than myself that was necessary for this book. And I greatly appreciate all their time and efforts and support and those were members of the political party's lab, as well as people that I hired with the NSF funding to and other sources of funding to read the statutes and help me understand how parties worked. What we're doing now is we're collecting party constitutions and statutes from around the world, not only in the countries that are studied in my book, but also in other cases. And we're making those publicly available so that people can have access to party rules and then hopefully can use more advanced techniques to quickly find out more about how parties function in a cross national investigation. And so using new. Sorry, using new tools, we can find out how parties select their candidates, their leaders and their manifesto, but also many other things like how they allocate financial resources, how people become members in the first place, how members can be removed from their position, what when meetings are held, and so on. And I collected a lot of that data, but only for a single period in time, so I don't have longitudinal data on these other variables. And so right now we're basically creating this repository where anybody can submit statutes and they become publicly available.
- [46:46]
B
That sounds fascinating and I'm really looking forward to seeing the work that comes out of the lab. So my guest today has been Georgia Cannell, Associate professor at the Department of Communication and Political Science at ucla. Her book Inside Parties is an exceptionally rigorous and data rich study of internal party rules and their consequences for party organization. Beyond providing one of the most comprehensive and comparative data sets on party organization, it makes an important theoretical contribution to our understanding of decentralization, internal democracy and political representation. I would specially recommend it to anyone interested in political parties, comparative politics, or mixed method research. It is a book that not only offers a substantive insight into party organization, but also serves as an excellent example of how to combine different research methods to answer complex questions. Georgia, congratulations once again on the book and thank you so much for joining me today.
- [47:51]
A
Thank you so much. Thank you for listening to this episode of the New Books Network. We are an academic podcast network with the mission of public education. If you liked this episode, please share it with a friend and rate us on your preferred podcast platform. You can browse all of our episodes on our website, newbooksnetwork.com Connect with us on Instagram and Bluesky with the handle EW Books Network and subscribe to our weekly substack newsletter@newbooksnetwork.substack.com to get episode recommendations straight to your inbox.