Political Insiders Outside The Party – An analysis of Irish ex-party independents and their transfer relationships with their former parties: 1982 – 2020.

1 MASTERS THESIS

Leiden University 2019/2020

Msc. Political Science - Parties, Parliaments & Democracy Name: Nils Fischer-Kerrane

Student no: S2660725 Submission date: 15.06.2020 Supervisor: Dr. Marijn Nagtzaam

2 Political Insiders Outside The Party – An analysis of Irish ex-party independents and their transfer relationships with their former parties: 1982 – 2020.

ABSTRACT: Over recent decades, independent (non-party) candidates have become the largest political grouping in Irish politics. Within this grouping, one subsection that seems to dominate are the independents who have previously been party candidates, a.k.a. ex-party independents. The electoral system in Ireland has a big role to play in this. While in most electoral systems, leaving a political party (and not joining another) usually signals the end of one’s career, this is not necessarily the case in PR-STV systems, where ex-party independents can survive and sometimes even thrive outside of the party system. In the elections following their departure, these candidates can win large amounts of votes and transfers, with a

significant number of those coming from their ex-party. This paper examines the relationships between ex-party independents and their former parties, focusing on the transfers that these candidate receives from that party. Furthermore, it aims to add to the study of ex-party independent candidates in Ireland, focusing on issues like electability, career longevity, origin parties, etc. and compares the performance of ex-party independents with those of other independent candidates.

Introduction

In 2010 Mattie McGrath, a Fianna Fail TD (Member of the Irish Dail), left his party and became an independent candidate. Despite winning less first preference votes in the subsequent election, McGrath managed to retain his seat in parliament. This was thanks in large part to the transfers he received from his former party, whom he would also end up

3 finishing ahead of. In discussing independent politicians around the world, Brancati observes how they “use their positions as outsiders to portray themselves as the only ones who can purge the government of corruption’’, and yet at the same time, so many are actually “political insiders’’ who began their political careers as part of a party (2008, 65). The Irish political system is a prime example of this, a country which despite having one of the smallest parliaments in western Europe, has elected more independents than the rest of Europe combined (Bolleyer & Weeks, 2009; 301). This has allowed the likes of McGrath to be reaping the benefits that were usually reserved for insiders. As a result, they are able not just to survive outside the party, but to thrive. The question is how many others like McGrath are there? To what extent do ex-party independent candidates benefit from their previous association with a party? And do they enjoy advantages that other independent candidates do not?

On party defections, Tavits observes that “legislators who have strong personal reputations are more independent of their parties’’ and that “defecting from the party line of voting is a manifestation of that independence’’ (2009; 796). While in other countries, such defectors would likely be forced to join another party, in Ireland they can easily transition to being independents who are well able to survive outside the party machinery (Weeks, 2011; 24). The example of McGrath suggests that the electoral cost of leaving the party was low (if not completely non-existent), and running as an ex-party independent even potentially benefitted his vote-seeking capabilities. Why does Ireland have so many independents? One of the primary reasons why independent candidates have been so successful in Ireland is thanks to the PR-STV system of voting (Weeks, 2014; Chubb, 1957). Mair has observed that that the electoral system is crucial as it allows voters to still cast a preference for such a defector without having to desert their preferred party (1987; 67). For ex-party independents this

4 means that that voters can still choose their usual party, while simultaneously assisting the electoral chances of a candidate whom they may have voted for in the previous election.

An Unknown Entity

The last decades have seen a significant growth in the number of independent candidates, meaning they now vastly outnumber any other political party or grouping on the ballot paper. While on the ballot paper, all such candidates are given the label of ‘’Non-Party’’, in reality there are multiple different types of independents (Darcy & Marsh, 1994; 48). It has been observed that as many as 70% of elected independents have a previous party affiliation (Weeks, 2011; 24), and yet it remains relatively unclear what role, if any, these affiliations play in the electoral success of these politicians. Despite being the most successful type of non-party politicians, ex-part independents remain a relatively unknown entity.

Weeks, who has led the way in this area, has previously examined the backgrounds of ex-party independents insofar as what their previous relationship with a ex-party was – i.e. if they were mere members, activists, elected officials, etc. (Weeks, 2011; 25). However he has not examined the electoral impact of this affiliation. Weeks also only went so far as to research whether these candidates were elected or not (Weeks, 2011; 25). He did not analyse any specific electoral data relating to these candidates – e.g. their ability to retain votes and transfers, win new votes and transfer, finish ahead of their former party, etc. More

significantly, there was no research into the relationship between the ex-party independents and the party with whom they began their political careers.The intention of this paper is to find all independents who previously stood for election as party candidates – in general elections or in local council elections between 1982 and 2020. And given that most ex-party independents end up running against candidates from their former party, we will be able to track the rate of transfers that pass from party to ex-candidate.

5 Any research in this area has been limited to those individuals who were already elected TDs when they left their party (Weeks, 2011; Chubb, 1957), i.e. those who succeeded in winning election. No research has been done into the many unsuccessful general election candidates, who began their careers as party politicians before going on to become independent. While they may have failed to win a seat, the electoral data of these individuals can offer an insight into some of the electoral consequences associated with becoming an ex-party independent, e.g. the impact on electability, no. of first preference votes and transfers, transfer ratio, voter retention, etc.

Transfer Solidarity & Hangovers of Loyalty

In Ireland, multi-member constituencies mean that often two or more candidates from the same party run against one another (Marsh, 2007; 504). As a result, parties hope for high levels of “transfer solidarity’’ between running mates, i.e. the percentage of a candidate’s transfers that go to a candidate from the same party (Gallagher, 1978; 2); e.g. if a candidate has 300 transfers to give and 150 of those go to their party running mate, then the this will mean a transfer solidarity rate of 50%. While transfer solidarity has been declining steadily for decades, it is still highest among the larger parties (Gallagher, 2005; 524). Of all political groupings, independent candidates have the lowest rate of transfer solidarity – 24%, while on the other hand, for parties it has been known to be as high as 82% (Gallagher 1978; 4). However this all we know about transfer solidarity among the broad spectrum of independent candidates. This paper aims to test what the transfer solidarity is among ex-party

independents.

Given the importance of the personal vote in Ireland, as well as the candidate-centred nature of the electoral system (Marsh, 2007; Martin, 2010), it is perhaps no surprise that some established party politicians can survive and even thrive outside of the party system. Weeks

6 (2011; 29) has identified a so-called ‘’hangover of loyalty’’ from previous elections, referring to the ability of candidates to hold onto a section of their previous voters, despite having left the party. By conducting voter surveys, Weeks found that many still gave a preference to a particular candidate, even if they are no longer standing for the same party (Weeks, 2011; 29-30). His research did not go beyond these voter surveys to look into electoral data. There has been no research into the data around the transfer solidarity between a party’s voters and independents previously of that party. Both Gallagher’s transfer solidarity theory and Weeks’ hangover of loyalty theory will be central to this paper, as we attempt to gauge the extent to which ex-party independents can benefit from vote transfers from their former parties.

Electability – “Pure” Independents vs. Ex-Party Independents

The vote-seeking priorities of politicians (Strom, 1990; 566) suggests that many will leave a party if they fear that remaining in it will negatively impact their election chances. Politicians tend to view strategy and career choices as ‘’trade-offs under risk’’ (Strom, 1997; 160), and will only choose party membership over independence if the potential benefits clearly

outweigh the negatives (Moser, 1999; 149-50). Considering that approximately 33% of party candidates lose out on a seat to a party colleague (Gallagher, 2005; 524), becoming an ex-party independent could increase the likelihood of winning a seat. This paper aims to investigate that.

Studies of independent politicians, both in Ireland as well as elsewhere, have shown how difficult it can be to be a successful ‘pure’ independent without any previous party affiliation (Weeks, 2009; Singleton, 2006). Weeks has looked at the backgrounds of successfully elected independents in Ireland and has found that approximately 7 out of every 10 have a

previous party affiliations (2011; 24). Of these, approximately 66% were elected as members

7 councillors before switching to independent (Weeks, 2007; 19). And although has argued that “independents are a particular type or phenomenon, which means they can be analysed from a comparative perspective” (Weeks, 2011; 41). Despite this there is yet to be a comparison made between the electoral performances of ex-party candidates and those of other

independents.

Analysing constituencies on an individual basis is also integral to any study of vote transfers (Endersby & Towle, 2014; 147), and Marsh remarks that “while lip service is frequently paid to the idea that there are 40 separate contests (one in each constituency), too often the

significance of this is ignored when commentators interpret the decision of the electorate.” (Marsh, 2007; 522). The strong localist aspects and candidate-centred nature of the Irish political system (Marsh, 1981; Weeks, 2009; Gallagher, 1985), make it important to look at independent candidates on a case-by-case (constituency-by-constituency) level. This allows for a comparison between the performance of ex-party independents and other independents, in order to examine whether being an ex-party independent gives an electoral advantage over other independents. This paper will look at each constituency in every election between 1982 and 2020, in order to see how ex-party independents perform against other independents.

PR-STV - The Irish Electoral System

Before discussing the research hypotheses and methods, it is important to outline exactly how the Irish electoral system works. In Ireland each seat is decided at a constituency level, with between approximately 40 constituencies electing a total of 160 members of parliament (TDs). District magnitude ranges from 3 to 5 seats in every constituency. Candidates are listed alphabetically on the ballot paper and not by party. Party candidates have the party name and logo next to theirs, with all other candidates given the label of ‘Non-Party’. Voters choose their preferred candidate by writing a ‘1’ in the box next to their name and photo, and

8 without a ‘1’ the ballot is deemed to be invalid (Doron & Kronick, 1977; 303). Once a first preference has been marked, voters are ‘’free, but not obliged’’ to rank (2nd, 3rd, 4th, etc.) as many of rest of the candidates as they wish (Gallagher, 1986; 254).

For each constituency, there is a different electroal a quota, which a candidate must reach if they wish to be elected. Alternatively they can also win a seat without reaching the quota, in the case of there being only two candidates left in the race and their total votes outnumber those of the other candidate’s (Endersby & Towle, 2014; 147). The quota for each

constituency is calculated as such:

If a candidate exceeds the quota and is elected, their surplus is divided up proportionally according to how that candidate’s second preferences are marked (Gallagher, 1986; 254). Once any surpluses have been distributed, the candidate with the lowest number of votes is eliminated, with their votes are ‘’transferred to the other candidates, according to the highest preference marked for any of the continuing candidates’’ (Gallagher, 1986; 254). Following this, the candidate with the second least amount of votes is eliminated and their votes distributed the exact same way. This process continues until all seats have been filled, either by candidates reaching the quota or there being no other candidates left to eliminate.

To give an example, if a constituency has an electorate of 100,000 and a district magnitude of 4, the quota in this case would be 20,001 votes. This means that if a candidate receives 10,000 first preference votes, they will need another 10,000 vote transfers in order to reach the quota and get elected. It is worth noting that only ever a small number of candidates are elected on first preferences alone, and therefore vote transfers are crucial in determining the vast majority of seats (Gallagher, 1978; 22, Weeks, 2010; 492).

Quota = 𝑻𝒐𝒕𝒂𝒍 𝒏𝒐.𝒐𝒇 𝒗𝒐𝒕𝒆𝒓𝒔

𝑻𝒐𝒕𝒂𝒍 𝒏𝒐.𝒐𝒇 𝒔𝒆𝒂𝒕𝒔+𝟏 + 1 (Doran & Kronick, 1977; 304)

9 Hypotheses

According to Sinnott & McBride (2011; 205-6), when analysing vote transfers there are two primary behavioural variables that need to be considered. These are 1) loyalty, i.e. the rate of transfers that go directly to a running mate, and 2) ‘’plumping’’, which Gallagher defines as “the willingness of supporters of that party to pass on transfers to candidates of other parties when their own party cannot make use of them’’ (Gallagher, 1978; 3). This means calculating the amount of ‘‘inter-party transfers’’, which emerge when ‘’there is no running mate

available to receive transfers but there is a continuing candidate of another party that is or has been in actual or potential, explicit or implicit alliance with the party of the candidate whose surplus or elimination votes are being distributed’’ (Sinnott & McBride, 2011; 206). These two variables – loyalty and inter-party transfers - will be crucial in shaping our research method as well.

Given that the candidates in this study will no longer be party members, the loyalty variable will be not exactly like Sinnott & McBride’s version but will also feature parts of Weeks’ ‘’hangover of loyalty’’ theory; the ability of ex-party independents to retain some voters after leaving a party (Weeks, 2011; 29). While Weeks focussed on first preferences and voter surveys, this paper will focus on lower preferences and electoral data.

Past research has shown that independent candidates are generally more ‘transfer friendly’ than party candidates, i.e. able to attract more inter-party transfers compared to party

candidates (Bolleyer & Weeks, 2009; 311). This means that as long as an independent can get a sufficient amount of first preferences to stay in the race, then as lower candidates are

10 them to members of the opposite party.” (Meredith, 1913, as cited in Gallagher, 2005; 523). It is expected that ex-party independents will be no exception to this general rule.

H1.1: Ex-party independents will receive more transfers as independents than they did as

party candidates, i.e. they will become more transfer friendly as independents.

Becoming more transfer friendly generally means becoming more transfer friendly to other independent voters as well (Weeks, 2011; 35). And while the levels of transfer solidarity among independents are relatively low (Gallagher, 1978; 3), we still expect ex-party

independents to attract some transfers from other independents at a higher rate than they did as party candidates.

H1.2: Ex-party independents will receive more transfers from other independents than they

did as a party candidate.

In Ireland, successful independents tend not to finish top of the poll but are much more likely to take the last seat on offer in a constituency. This happens not by attracting significant numbers of first preferences, but instead by staying in the race and attracting plenty of vote transfers (Meredith, 1913, as cited in Gallagher, 2005; 523, Chubb, 1957; 135, Weeks, 2014; 607, Gallagher, 1986; 271). This seems to show that independents tend to receive less first preferences than party candidates, but significantly more transfers. We also expect ex-party independents to see a reduction in the number of first preferences they receive – compared to the number received as a party candidate.

11 H1.3: Ex-party independents will receive less first preferences than they did as party

candidates

In their description of inter-party transfers, Sinnott and McBride refer to a ‘’potential, explicit or implicit alliance’’ between candidates of different parties (2011; 206). Combine this with the fact that voter surveys have also shown the presence of a ‘’hangover of loyalty” for ex-party independents (Weeks, 2011; 29), we expect the transfer relationship between a ex-party and its ex-candidate to be stronger than that between candidates of different parties. In other words, an ex-party independent candidate is likely to receive the highest share of inter-party transfers from his/her former party, compared to the shares received by all other candidates in that constituency.

H2: Ex-party independents will be the biggest recipients of interparty transfers from their

former party.

Declining levels of transfer solidarity resulting in ‘’lukewarm intra-party relationships’’ (Marsh & Sinnott, 1997; 158-9) make life outside the party seem more and more attractive. Given that personal vote seeking behaviour (like exiting a party) is prompted to a large extent by intra-party competition (Carey & Shugart, 1995; 177, Selb & Lutz, 2015; 335), we expect that a candidate will choose to leave a party and run against them if they are reasonably confident that they can finish above that party.

H3: In the election that immediately follows their departure, ex-party independents will finish ahead of candidates from their former party.

12 It is not just their former party colleagues that ex-party independents will hope to finish ahead of, they will also want to beat off any competition from other independents. Assuming that ex-party independents will become more transfer friendly to all voters (including those of their former party), we expect that the “quasi-party status’’ (Bolleyer & Weeks, 2009; 304) they enjoy should give them an advantage over their independent colleagues, and they are therefore likely to finish above them in the majority of constituencies.

H4: In constituencies with more than one independent candidate, one of which is an ex-party independent, the ex-party independent is expected to be the highest performing independent.

Singleton observes how independent politicians ‘’do not have to sacrifice principles to maximise their vote’’, and therefore ‘’adhering to the principles upon which they stood for election will reinforce the support of their constituency’’ (Singleton, 1996; 79). This freedom that comes with being an independent suggests that they have more freedom to focus on their personal vote, and therefore increase the likelihood of retaining votes. Given that the personal vote plays a central role in the Irish political system and consequently contributes to the success of independents (Marsh, 2007; Martin, 2010; Chubb, 1957; Weeks, 2014), this should allow ex-party independents to maintain or improve their transfer rate in their second election after leaving the party.

H5.1: In their second election as independents, ex-party independents will maintain or

13 While ex-party independents benefit from a ‘’hangover of loyalty’’ in their first election after leaving a party, voter surveys have shown that this does not go on indefinitely and tends to reduce as time passes (Weeks, 2011; 29). We expect the electoral results to reflect this.

H5.2: In their second election as independents, ex-party independents will receive less

inter-party transfers from their former inter-party than they did in the previous election

Methodology

This paper defines ‘ex-party independent’ as any non-party general election candidate who has previously contested an election – local or general – as a party candidate. All of these will

have been listed as “Non-Party” on the ballot paper (Darcy & Marsh, 1994; 44). By looking into the electoral background of each independent candidate, we have compiles a list of all

ex-party independents who have stood in general elections and/or local elections between

1982 and 2020.

The reason for this time period is not only because it will provide the most up to date information, but given that the last couple of elections have featured the largest number of non-party candidates, it will also allow for the largest sample of politicians. Furthermore, the election of February 1982 is the first election in recent history where there were more

independent candidates running than there were constituencies (Weeks, 2009; 27). Given that one of the aims of this paper is to compare ex-party independents with all other independents, the fact that independents outnumber the amount of constituencies, makes it very likely that there will be multiple constituencies with more than one independent candidate. This then allow for a comparison to be made.

14 All in all, this time period will cover a total of 11 elections, within which there will be 1099 independent candidates running in 454 constituencies. Data was organized into two different sets. The first contained all candidates who contested general elections, at least once as a party candidate and at least once as an independent, between 1982 and 2020. The data collected for each candidate was 1st preferences (as a percentage of the quota), transfer ratio, transfers received from former party, and transfers received from other independents. It was also noted whether the ex-party candidate finished ahead of candidates from their former party, as well as whether they succeeded in being elected – both as a party candidate and as an independent.

Here is an example of one such candidate and all of his relevant data:

Figure A: Example of data collected for ex-party general election candidate

The second dataset will be all ex-party independents between 1982 and 2020, who had previously contested local elections as party candidates, and who were now contesting their first ever general election. The data collected will be from their first general election as independents. We will take note of the no. of transfers received from their former party, whether they finished above their former party, as well as whether they were elected or not.

15 Here is an example of one such candidate and his relevant data:

Figure B: Example of data collected for ex-party local election candidate

The method for testing H4 - the rate at which ex-party independents finish ahead of other independents – will involve going through every election and every constituency between 1982 and 2020, identifying those where more than one independent is running, and then checking if at least one of those independents is an ex-party independent. We can then compare the number of constituencies in which an ex-party independent was the highest finishing independent, with all the constituencies in which he/she wasn’t the highest performing independent.

For example, here is the relevant information collected for the 2020 election:

Figure C: Example of constituency data

In this example, 17 (5 + 12) constituencies had either 1) no independents, 2) only one

independent, or 3) no ex-party independents. Of the remaining relevant 23 constituencies, 17 (74%) of them were constituencies in which an ex-party independent was the highest

16 As mentioned above, one of the variables for each candidate will be the transfer ratio. This is found by taking the number of transfers received and then calculating them as a proportion of their first preferences received. Here is an example taken from the dataset:

Figure D: Example of transfer ratio calculation

In this example we see that the 2016 transfer has been calculated by dividing 377 into 1743, giving us a transfer ratio of 0.21. The subsequent election saw the transfer ratio increase to 0.74. Overall, this means that this particular candidate increased their transfer ratio by 252% after becoming an ex-party independent.

Data sources

Data was collected from the two main electoral results databases; ElectionsIreland.com and

ElectionsIreland.org. The former provides a comprehensive breakdown constituency by

constituency of how votes and transfers were distributed. The latter is the best source for the

electoral history of individual candidates’, allowing us to see which independent candidates have previously contested elections as party members.

17 RESULTS

The 11 general elections that occurred between February 1982 and February 2020, saw 229 (20.8%) of the 1099 independent candidates running fit our description of ex-party

independent. When we subtract the campaigns of those who ran more than once as

independents, we end up with 116 individuals who we deem to be ex-party independents. Of those 116 individuals, exactly 50% (58) had previously contested a general election as a party candidate. The other 58 had previously run in local election(s) representing a party.

Figure 1: Total no. of independents and Total no. of ex-party independents

58 48 85 49 103 104 95 90 177 160 133 2 4 7 5 13 12 15 14 33 33 ₃₁ 1982 (Feb) 1982 (Nov) 1987 1989 1992 1997 2002 2007 2011 2016 2020 Total Independents Ex-party Independents

18 Figure 2 below displays the transfer ratios of all candidates who contested general elections as both a party candidate and an independent. Of the 58 candidates, 39 (67%) became more transfer friendly after becoming independent. and there was a mean increase of 21.9% in the transfer ratio across all 58 candidates. In other words, the mean transfer ratio as party

candidates was 64:100 i.e. 64 transfers received for every 100 1st preferences. Once candidates became independent however, the mean transfer ratio became 78:100. This confirms that H1.1. is as expected.

The next step was to breakdown the sources of the transfers, and to calculate what percentage came from other independent candidates. As we can see from Figure 3, ex-party independent candidates succeeded in becoming more transfer friendly to more independent voters. In total, 70% of candidates increased the amount of transfers they received from independent voters. As party candidates, these individuals had received an average of 10.7% of

independent transfers. Once they became independent, that figure went up to 14% - an increase of 31%. As a result, we can conclude that H1.2 is also as expected.

The bar chart outlining first preferences shows that, on average, the number of first

preferences received by candidates reduced by 44% after the candidate moved to being an independent. In their final election as a party member, candidates had a mean first preference ratio of 0.6. This means that if the mean quota was 10,000 votes, then the average number of first preferences received was 6000. Once the candidates moved to being ex-party

independents, the number of first preferences went down to an average of 0.34 of the quota. Therefore we can confirm that H1.3 is as expected.

19 Figure 2: Former party general election candidates as Ex-Party Independents – Transfer Ratios

0 0.5 1 1.5 2 2.5 3 N o la n M cN a m a ra Ly o n s B u rk e G o g a rt y C o ll in s D u ff G re a li sh H e a ly Le a h y Fi tz p a tr ic k Bree _Perr y M cG ra th B e h a n C o n la n B ro u g h a n P o w e r M a th e w s M a lo n e y M cG u in n e ss E a m e s Se xt o n C o w le y Fo le y H a ll ig a n D w ye r M cG a rv e y M cG ra th W e lb y Fl e m in g M u rp h y K e ll y C o x Fl yn n Fi tz g e ra ld K e ll y T o rm e y M a cD o w e ll K e n n e d y G re h a n C o n a g h a n H yn e s M a cG e a ra il t O F o ig h il B u tt e rf ie ld Belle w C o sg ro ve W il ki n so n M a rk e y G le n n Sk e ll y T re a cy B ro p h y O 'C a ll a gh a n TR A N SF ER R A TI O

As party candidate As independent Mean transfer ratio as party Mean transfer ratio as indepedent

0,78

20 0 10 20 30 40 50 60 70 N o la n Ly o ns B ur ke B u rk e G o gar ty G o gar ty C o lli ns Le ahy Fi tz pa tr ic k N au ght en B re e B re e M cG ra th M cG ra th M cG ra th B eha n B eha n B eha n P o w er Fo le y Fo le y H al lig an M cG ar ve y Fl em ing _Cox Fl ynn Fi tz ge ral d Ke lly To rm ey Ke nn edy M ac G ear ai lt B ut te rf ie ld C o sg ro ve M ar ke y G le nn Sk el ly Tr ea cy C o nn o lly C o nn o lly C o nn o lly D al y O 'Do no ghu e O 'Do no ghu e M o ran M o ran M ur ph y Snel l Ke o ha ne P ri ng le P ri ng le P ri ng le B ar ry B re en B re en G le eso n Ki el y C ri lly M cC ar thy R o ch e Ta af fe O 'Le ar y H ur le y Fo x B la n ey D ur ca n M ar ry % o f ex -p ar ty t ra n sf ers re ce iv ed 16,7 Mean %

21 Figure 4: Transfers from Independents – As independent and as party candidate

0 10 20 30 40 50 60 Perc en ta ge o f o ve ra ll tra n sf ers (% )

As party candidate As independent

10,7 14

22 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 Lo w ry N o la n M cN a m a ra Ly o n s B u rk e G o g a rt y R o ss C o ll in s D u ff G re a li sh H e a ly Le a h y Fi tz p a tr ic k N a u g h te n Bree _Perr y M cG ra th B e h a n C o n la n B ro u g h a n P o w e r M a th e w s M a lo n e y M cG u in n e ss E a m e s Se xt o n C o w le y Fo le y H a ll ig a n D w ye r M cG a rv e y M cG ra th W e lb y Fl e m in g M u rp h y K e ll y C o x Fl yn n Fi tz g e ra ld K e ll y T o rm e y M a cD o w e ll K e n n e d y G re h a n C o n a g h a n H yn e s M a cG e a ra il t O F o ig h il B u tt e rf ie ld B e ll e w C o sg ro ve W il ki n so n M a rk e y G le n n Sk e ll y T re a cy B ro p h y O 'C a ll a gh a n % O F QU O TA ( 1 = 100 % O F QU O TA )

As party candidate As independent

0,34

Mean 1st preferences as party candidate Mean 1st preferences as independent

23 Of the 229 ex-party independent campaigns that occurred between 1982 and 2020, there were a total of 66 occasions in which an ex-party independent was in a position to receive transfers from their former party.There are four reasons why the other campaigns could not be

included, which are 1) the candidate was elected before transfers from their former party became available, 2) the candidate was eliminated before transfers from their former party became available, 3) the former party did not have a candidate running in the same

constituency as a former party independent, and so no transfers from that party were available, or 4) the former party of the ex-party independent has disbanded and no longer exists – also known as ‘’vestigial independents’’ (Weeks, 2009; 9).

The ex-party transfers received by these 66 candidates have been compiled in the table below (Figure 5). Included as well is the mean percentage of transfers received (16,7%), which is displayed as a dotted broken line.

Of those 66 candidates who were in a position to benefit from transfers from their former party, 25 (38%) of them were the biggest transfer recipients, i.e. they received the highest share of transfers from that party’s candidate(s). The remaining 41 (62%) were not the biggest recipients, meaning that another candidate in their constituency received the highest number of inter-party transfers from the party in question. As a result, we can conclude that H2 is not as we had expected - the majority of ex-party independents were not the main recipients of transfers from their respective former parties.

24 Figure 6: Percentage of Ex-Party Independents who were main recipients of inter-party

transfers from their former party

Any time a political candidate leaves a party or is forced to leave a party, they will ultimately hope that the electoral cost is not so high. The may even hope that it could improve their electoral prospects. Of the 229 ex-party independent campaigns researched in our sample, 156 involved an ex-party independent running against a candidate from their former party. The reason for not including all campaigns is that there were some cases where the former party in question chose not to stand a candidate in that constituency, as well as the fact that there were also cases where the ex-party no longer existed and so therefore had no candidates competing in any constituency. In those 156 campaigns, the ex-party independent finished ahead of their former party on 59 (37,8%) occasions. This means that the remaining 97 times, the ex-party independent finished behind all candidates representing their ex-party. Therefore we must conclude that H3 is not as we expected.

38%

62%

Ex-party Independent = Main Recipient of inter-party transfers Another candidate = Main recipient of inter-party transfers

25 Figure 7: Percentage of ex-party independents who finished ahead of their former party in elections following their departure

It is worth noting that this section of the research demonstrated the major distinction between the candidates who had contested a general election before and those who had not. Of the ex-party candidates who had previously contested general elections as ex-party candidates, a majority (51%) of independents managed to finish ahead of candidates from their former party. However the number of ex-party councillors who finished ahead of their former party is considerably lower (23.5%).

37,8%

62.2%

26 Figure 8a: Ex-party general election candidates Figure 8b: Ex-party local election candidates

Our fourth hypotheses (H4) relates to the rate at which ex-party independents finish above other independent candidates, a.k.a. ‘’pure’’ independents. Over the 11 elections and 454 constituencies examined in this study, 127 of them were constituencies in which ex-party independents and other independents competed against one another for votes. Closer analysis of those 127 constituencies shows that in 87 of them, an ex-party independent was the highest performing independent candidate. This works out as 69% of constituencies and confirms the hypotheses that, in the majority of constituencies, ex-party independents will be the highest performing independents.

51%

49%

Finished ahead of former party Finished behind former party

23.5%

76.5%

Finished ahead of former party Finished behind former party

27 Figure 9: Percentage of constituencies with ex-party independents with other independents AND Percentage of constituencies where ex-party independent as highest independent

As the above chart demonstrates, the steady rise in constituencies where two different types of independents compete is matched by the rise in the number of constituencies wherein ex-party independents finish as the highest independent. The figure of 69% of relevant

constituencies becomes even more significant when we consider that ex-party independents make up just 13% of the total number of independent candidates who stood in those eleven elections. So despite making up only a fraction of the total independent candidates, ex-party independents are likely to exert a lot of electoral dominance and outperform their

independent colleagues. 0.00% 10.00% 20.00% 30.00% 40.00% 50.00% 60.00% 70.00% Feb 1982 Nov 1982 1987 1989 1992 1997 2002 2007 2011 2016 2020 % o f to ta l n u m b er o f co n stit u en cie s

Total constituencies with +1 IND (incl. 1 Ex-P IND) Constituencies with Ex-P IND as highest IND

28 Of the 116 ex-party independents in this study, 30 chose to contest a second election. Three of those reached the quota on first preferences alone, leaving us with 27 ex-party

independents who availed of transfers in their second election. As we can see from the mean, the average transfer ratio not only remained consistent, but considerably increased from 0.29 in the first election up to 0.49 in the second election. As a result, we can conclude that H5.1 is as expected.

69%

31%

Highest independent = Ex-Party Independent Highest Independent = Other independent

13%

87%

Ex-Party Independents All other independents

Figure 10a: Constituencies with more than 1 independent, including 1 ex-party

independent

Figure 10b: Total ex-party independents & Total of all other independents (1982 –

29 Figure 11: Transfer Ratio – 1st _{Election as Independent & 2}nd _{Election as Independent}

0.06 0.66 0.39 0.09 0.58 0.17 0.68 0.07 0.07 0.41 0.19 0.52 0.11 0.34 0.23 0 0.31 0.43 0.29 0.74 0.14 0.3 0.08 0.39 0.03 0.22 0.29 0.52 0.83 1.75 0.7 0.76 0.7 0.08 0.28 0.04 0.29 0.14 1.21 0.29 0.91 0.47 0.05 0.53 0.15 0.12 0.62 0.83 0.25 0.04 0.1 0 0.95 0.49 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Tra n sf er ra tio

1st election as independent 2nd election as independent 0,49

0,29

30 Regarding inter-party transfers in the second election, not all ex-party independents who contested a second election were in a position to receive inter-party transfers from their former party. Just like with the first election, there are multiple reasons for this: 1) candidate was elected before transfers from their former party became available, 2) candidate was eliminated before transfers from former party became available, 3) former party did not have a candidate running in the same constituency, or 4) the former party of the ex-party

independent has disbanded. In total, there were 11 ex-party independents who were in a position to receive transfers from their former party. As we can see from the table below, there was an overall decrease in the mean number of votes between the first election as independents and the second election. The mean rate of transfers that came from the ex-party went from an average of 22.3% down to 16.1% - a decrease of 28%. Hence, this allows us to conclude that H5.2 is as expected.

Figure 12: Time effect of ex-party transfers – 1st _{election as independent & 2}nd _{election as independent}

18 26 11 43 7 5 52 14 14 34 21 22.3 37 29 14 22 8 3 17 11 17 3 16 16.1 0 10 20 30 40 50 60 % o f tra n sf ers f ro m f o rm er p ar ty

Time Effect of Ex-Party Transfers:

1st election as independent vs 2nd election as independent

31 Given that the intention of this paper is also to add to our overall understanding of ex-party independents, the collected data reveals three other significant variables. These are

electability, career longevity, and parties of origin. Each of these three will now be briefly discussed.

Of the 116 candidates in their first election as ex-party independents, 22 (19%) of them succeeded in being elected or re-elected (in the case of already sitting TDs). This means that the other 94 (81%) failed to win a seat or lost their seat (in the case of already sitting TDs).

Figure 13 : Percentage of ex-party independents elected

If we breakdown these figures between the list of candidates who had previously contested a general election and those who only previously contested local elections, we can see some significant differences in the electoral fates of the candidates. Of the sitting TDs, the same number lost their seats as independents as the number who gained a seat – 5. Twice as many candidates (10) succeeded in retaining the seats that they had originally won as party

members. However of all the candidates who contested general elections as party members

19%

81%

32 and then ex-party members, the largest majority (66%) still failed to win a seat on both occasions. Of the ex-party councillors who contested general elections between 1982-2020, only a fraction manage to ever win a seat. 12% of ex-party councillors succeed in winning a seat in their first attempt in a general election, while a further 7% of ex-party councillors fail on their first attempt but then succeed at a later general election.

Figure 14: Previous general election candidates becoming ex-party independent – Impact on Electability

Figure 15: Previous local election candidates contesting general elections as ex–party independents

66% 17%

8,5% 8,5%

Never elected Retained seat as independent Lost seat as independent Won seat as independent

0% 10% 20% 30% 40% 50% 60% 70% 81% 7% 12% Never elected Elected on 2nd/3rd attempt Elected on 1st attempt

33 Another significant statistic relates to the career longevity of ex-party independent, which is shown in the overall figure for the amount of candidates choosing to contest a second election at all. We can see that from the first election as ex-party independent to the second election, there is a notable reduction in the number of ex-party independents. This indicates that many ex-party independents choose only to contest one election as independents.

Given that our research is focussed on candidates contesting back-to-back elections, we need to exclude those independents who ran in 2020, and are therefore still likely to compete in the next election. However of the other ex-party independents who contested elections between February 1982 and 2016, we found that these candidates ran in a total of 406 elections (local and general) as party candidates. Once becoming independent, they would only end up

contesting a total of 200. This roughly works out as a ratio of 2:1, meaning that for every two elections a candidate contested as a party candidate, they only contested one as an ex-party independent.

Figure 16: Total elections as party candidates & Total elections as ex-party independents

406

200

34 The fact that candidates contest twice as many elections as party members than as

independents shows the reliance that many have on the party system and the difficulties associated with running and maintaining independent campaigns. These difficulties could be to do with a lack of finances, a lack of coverage, access to polling information, etc.

Figure 17: Ex-Party Independents – Parties of Origin

The final piece of significant date relates to the parties of origin of ex-party independents. The pie chart above breaks down the ex-party independents party-by-party. Among the 10 parties who had ex-party independents standing in elections between 1982 and 2010, an average of 3,17% of former candidates went onto become ex-party candidates. This means that for every 100 candidates a party runs over multiple elections, approximately 3 former party candidates will be running as independents. Of all the parties, the Democratic Socialists had the highest rate of former party candidates running as independents. However they are the outlier in the list of parties, most likely to with the fact that the party disbanded prior to

27% 28% 7% 22% 4% 4%4% 1% 1% 1% 1%

Ex-Party Independents - By party

Fine Gael Fianna Fail Sinn Fein Labour

Green Party Progressive Democrats People Before Profit Renua Workers' Party Democratic Socialists Democratic Left

35 the 1992 election (Coakley, 1990; 284). Had they continued to exist beyond that election and continued to run candidates, the rate of ex-party independents to party candidates would have been reduced. By 1992, the Democratic Socialists had evolved into Democratic Left

(Dunphy, 2007; 55), whose percentage of 3,0% is much closer to the mean.

Figure 18: Party By Party – Total candidates & Former Party Independents

TOTAL NO. OF CANDIDATES TOTAL EX-PARTY INDEPENDENTS

%

Over the course of the 11 elections included in this research, only three parties (Fianna Fail, Fine Gael, Labour) have contested every one of those election. Additionally, Sinn Fein,

36 Greens and the Workers’ Party contested a majority of the 11 elections. Of all these parties who contested most of the elections between 1982 and 2020, the Labour Party has a significantly higher rate of ex-party independents – 5,3%. The reasons for this are likely related to both internal issues within the party as well as external issues within the wider Irish political environment. For decades, the Labour Party has experienced high amounts of

internal division, which is then reflected in the high number of ex-Labour candidates who go onto become independent later in their political careers (Weeks, 2009; 16). The evolution of the Democratic Socialists into Democratic Left, which would then also eventually merge with the Labour Party also created a political vacuum on the left in Ireland which was then filled by independents, many of whom would have been previous member / candidates of these three parties (Weeks, 2009; 16). Ideological independents, one of several categories of non-party candidates in Irish politics, have tended to mostly have been ex-Labour members who quit the party due to its ‘’ non-confrontational conservative approach’’ (Weeks, 2009; 16). All of this combines to ensure that the Labour Party have the highest number of ex-party independents.

DISCUSSION

The last decade of Irish elections has seen the highest number of independent candidates on ballot papers. 2016 was not only the year with the most amount of non-party candidates, but also the highest number ever elected – 23 in total. This proved to be not only a national record, but was also internationally significant as it was the highest percentage of

independents in any national legislature of any ‘’mainstream democracy’’ anywhere since 1950 (Weeks, 2016; 207). The findings of this paper show that 2016 (alongside 2011) was also the year with the highest no. of ex-party independents. This suggests that as the number

37 of independents rises so too will the number of ex-party independents, demonstrating the importance of researching this ever-growing political grouping.

One of the primary goals of this paper was to highlight the transfer relationship between ex-party independents and their former parties. Not only have we added to the understanding of that relationship, but we have also quantified it with exact data and figures of transfers that go from an ex-party to an ex-candidate. Furthermore, we have added to the relatively unknown phenomenon that is the ex-party independent, by exploring issues like transfer friendliness, transfer solidarity, plumping, electability, career longevity, as well as parties of origin. While in many other electoral systems such a relationship between a former party candidate and that party’s voters is invisible and unquantifiable, the Irish system of PR-STV has allowed us to demonstrate the presence of such a relationship. Overall, we have seen that a large number of ex-party independents can benefit from this relationship, and there are only a handful of cases where the departure from the party has a negative impact on the electoral performance.

Of all the data presented in this paper, perhaps the most notable is the data on the percentage of constituencies in which ex-party independents were the highest performing independents. Despite only making up for a small fraction (13%) of the total independents, these ex-party candidates seemingly dominate when it comes to winning independent votes and transfers. This is demonstrated by the fact that when they go head-to-head with other independents, they finish ahead of them 69% of the time. These figures go some way to suggesting that when it comes to being an independent candidate, a previous party affiliation can prove to be very electorally beneficial. If, as Brancati (2008; 650) suggests, independents are either political insiders departed from a party or else political outsiders, it is the former who are most likely to succeed in any run-off between the two.

38 The nature of the STV system of voting allows for in-depth research into the subject of vote transfer solidarity, i.e. the number of transfers received from candidates of the same party. Non-party candidates have the lowest rate of transfer solidarity, considerably lower than the rate of transfers between candidates of the same party, and even candidates of different parties. The figure below show that the transfer solidarity rate between independents and non-party independents is 14%. While ex-non-party independents still attract more independent transfers than they did as party members (10.7%), this number is still significantly lower than the historical average of independent transfer solidarity – 24% (Gallagher 1978; 4). It is also considerably lower than the most recent figures on transfer solidarity among the major parties, which averages out at 63% (Gallagher, 2005; 524-5). This suggests that for some independent voters, ex-party independents are not as attractive as other independents and therefore less likely to benefit from transfers from these voters. Given that for many voters, independent candidates are an anti-party vote (Belanger, 2004; 1057, Weeks, 2009; 21, Weeks, 2011; 35), the fact that ex-party independents have a previous party affiliation likely means that a certain percentage of independent voters will not vote for them, either with a first preference or a transfer further down the ballot sheet

If we are to take the view that the goal of many politicians are primarily vote-seeking (Strom, 1990; Shugert et al., 2005; Selb & Lutz, 2015), then it would seem that the electoral cost of leaving a party to become independent is relatively low. While many of candidates ultimately still failed to win a seat, only a handful suffered electorally as a result of becoming

independent. For the majority, it proved to be a political manoeuvre that brought them at least somewhat closer to taking a seat in parliament.

However, the ratio of total number of elections contested as ex-party independents to the total number contested as party members is a very telling statistic. The fact that candidates contest twice as many elections as party members, demonstrates the pivotal role that a party can play

39 in both the survival and success of a candidate. Norris has argued that in most democracies, the chances of survival for non-party politicians ‘’without the official endorsement, financial

assistance, and organisational resources that parties provide’’ is very low (Norris, 2006; 91). The data here goes some way to confirming that this is true for ex-party independents. While

in the short term it seems that surviving as an independent in Ireland is possible, career

longevity is much more likely inside a party than outside it.

As party loyalty declines, independent candidates become an attractive choice to voters as they do not require them to completely change their previous party allegiance (Weeks, 2011; 40). Ex-party independents can greatly benefit from this, particularly in this age of the floating voter, where candidates may not necessarily need to define themselves as who they are, as long as they can make clear who they are not. All independent candidates have the ability to attract anti-party and anti-system votes (Weeks, 2011; 38), and the fact that many independent voters are not frustrated at all parties but just one specific party (Weeks, 2011; 20), means that ex-party independents can fill a political vacuum. In the same way that non-partisan voters ‘’use their ballot to hold parties accountable for past performance’’

(Dassonneville et al., 2015; 401), the same voters can also express their dissatisfaction with a party by giving their vote to the ex-party independent recently departed from said party. Furthermore as this paper has demonstrated, they can also attract significant vote transfers from their former party. This ‘best of both worlds’ scenario allows these politicians to

survive, and occasionally even thrive, outside of the party with whom they originally became political insiders.

As political parties around the globe are losing their ability to act as ‘’vehicles of representation, instruments of mobilization, and channels of interest articulation and aggregation’’, one of the few areas where they have maintained control is with the

40 recruitment of election candidates (Biezen, 2004; 1). At the same time, the increasing number of ex-party independents in Ireland shows that while the majority of political candidates still begin their careers within parties, keeping them within the party can be quite challenging. If it is possible to survive outside of a party while simultaneously benefitting from one’s

connection to that party, it is likely that the number of ex-party independents will continue to rise.

The candidates included in this study were those who had contested elections between 1982 and 2020, and our criteria for assessing an ex-party status based on whether an individual had stood for a party, locally or nationally, during that period. Future research could be expanded to include independents who had previously been members of a party without ever having contested any elections. This would involve more extensive research into the backgrounds of candidates. By examining the transfer relationship between an ex-party member and their former party, it could then compare that relationship with the one addressed in this paper.

The aim of this paper has not been to quantify the exact impact that leaving a party and becoming independent has on a candidate’s electability. There are many factors that influence a candidate’s election results, and comparing the results of one election with another has its limitations. There is no way of knowing just how significant (or insignificant) the switch to being independent had on the overall vote share, and what other factors were involved. Given the Irish electorate’s tendency to engage largely in retrospective voting, an ex-party candidate will often win or lose votes and transfers based on their ability to deliver as a TD, local councillor, etc. (Marsh, 2007; 509). Furthermore, approval ratings for non-party candidates, just like with any party, fluctuate significantly from one election to another (Weeks, 2011; 24). Multiple factors influence the electoral performances of ex-party independents and to account for all of them would require much more research.

41 During the collection and analysis of the data, something that became clear is the fact that the transfer relationship between ex-party and ex-candidate is a two way relationship, This means that, as well as receiving large amounts of transfers from their former parties, many ex-party independents also provide a large amount of transfers to the party. In many cases, they even provide more transfers to their former party than the party does to them. Examining the extent to which parties benefit from transfers from their ex-party independents could be a potential area for future research.

Lastly, this research paper did not have the opportunity to examine the reason(s) why a candidate left a party. There are many reasons why an individual might leave a party – disagreements with other members, personal reasons, office-seeking motivations, improved career opportunities, voting against party, etc. (Laver & Benoit, 2003; 216; Bolleyer & Weeks, 2009; 307; Desposato, 2006; 74; Brancati, 2008; 650, Weeks, 2009; 16). The nature of a candidate’s departure from a party would inevitably impact their performance in a subsequent election. Future research could explore the different types of exit from a party, and whether certain types of exit have a greater impact on an ex-party independents ability to retain votes.

42 BIBLIOGRAPHY

Bélanger, É. (2004). ‘Antipartyism and third-party vote choice: A comparison of Canada, Britain, and Australia’, Comparative Political Studies, 37(9), 1054-1078.

Biezen, Ingrid van (2004). ‘How Political Parties Shape Democracy’. Irvine: Paper 04'16, Center for the Study of Democracy, University of California.

Bolleyer, N. & Weeks, L. (2009). ‘The puzzle of non-party actors in party democracy: Independents in Ireland Comparative European Politics’, 7(3), 299–324. Brancati, D. (2008). ‘Winning alone: The electoral fate of independent candidates

worldwide’ The Journal of Politics, 70(3), 648-662.

Carey, J. M., & Shugart, M. S. (1995). ‘Incentives to cultivate a personal vote: A rank ordering of electoral formulas’, Electoral Studies, 14, 417-39.

Chubb, B. (1957). ‘The independent member in Ireland’, Political Studies, 5(2), 131-139. Coakley, J. (1990). ‘Minor parties in Irish political life, 1922–1989’, Economic and Social

Review, 21(3), 269-97.

Darcy, R. & Marsh, M. (1994), ‘Decision heuristics: Ticket-splitting and the Irish voter’, Electoral Studies, 13(1), 38-49.

Dassonneville, R., Blais, A. & Dejaeghere, Y. (2015) ‘Staying With the Party, Switching or Exiting? A Comparative Analysis of Determinants of Party Switching and

Abstaining’, Journal of Elections, Public Opinion and Parties, 25(3), 387-405. Desposato, Scott W. (2006) ‘Parties for rent? Ambition, ideology, and party switching in

Brazil's chamber of deputies.’ American Journal of Political Science, 50(1), 62-80. Doron, G., & Kronick, R. (1977). ‘Single transferrable vote: An example of a perverse social

choice function’, American Journal of Political Science, 303-311. Dunphy, Richard (1998) ‘A group of individuals trying to do their best’: The

43 ElectionsIreland.com, Irish General Elections Results, Counts, Stats And Analysis. [online].

Available at: <http://irelandelection.com/elections.php?electype=1> [Accessed 14 June 2020].

ElectionsIreland.org, Elections Ireland: General Elections. [online] Available at:

<https://electionsireland.org/results/general/index.cfm> [Accessed 14 June 2020]. Endersby, J. W., & Towle, M. J. (2014).’Making wasted votes count: Turnout, transfers, and

preferential voting in practice’, Electoral Studies, 33, 144-152.

Gallagher, M. (1978). ‘Party solidarity, exclusivity and inter-party relationships in Ireland, 1922–1977: the evidence of transfers’, Economic and Social Review, 10(1), 1-22.

Gallagher, M. (1985). ‘Social backgrounds and local orientations of members of the Irish Dail’, Legislative Studies Quarterly, 10, 373-394.

Gallagher, M. (1986). ‘The political consequences of the electoral system in the Republic of

Ireland’, Electoral Studies, 5(3), 253-275.

Gallagher, M. (2005). ‘Ireland: the discreet charm of PR-STV’. In Gallagher, M., & Mitchell,

P. (Eds.) The politics of electoral systems. OUP Oxford, 511-532.

Laver, Michael, and Kenneth Benoit. (2003) "The evolution of party systems between elections." American Journal of Political Science, 47(2), 215-233.

Marsh, Michael (1981). ‘Localism, Candidate Selection and Electoral Preference in Ireland’, Economic and Social Review, 12, 167–86.

Marsh, Michael (2007). ‘Candidates or Parties? Objects of Electoral Choice in Ireland’, Party Politics, 13(4), 500–527.

Martin, Shane (2010) ‘Electoral Rewards for Personal Vote Cultivation under PR-STV’, West European Politics, 33(2), 369-380.

Martin, S. (2016). ‘Policy, Office and Votes: The Electoral Value of Ministerial Office’, British Journal of Political Science, 46(2), 281-296.

44 Mair, Peter (1987). ‘Do Parties Compete?’. In Mair, P. ‘The changing Irish party system:

Organisation, ideology and electoral competition’ (pp. 61-94), Burns & Oates. Moser, R. G. (1999). ‘Independents and party formation: elite partisanship as an intervening

variable in Russian politics’, Comparative Politics, 147-165.

Selb, P., & Lutz, G. (2015). ‘Lone fighters: Intraparty competition, interparty competition, and candidates' vote seeking efforts in open-ballot PR elections. Electoral Studies, 39, 329-337.

Singleton, Gwynneth. (1996). ‘Independents in a multi-party system: The experience of the Australian senate.’ Papers on Parliament 28, 61-82.

Sinnott, R., & McBride, J. (2011). ‘Preference voting under PR-STV 1948–2011’. In How Ireland Voted 2011 (pp. 205-221). Palgrave Macmillan, London.

Strøm, K. (1990). ‘A behavioral theory of competitive political parties’, American Journal of Political Science, 34(2), 565-598.

Strøm, K. (1997). ‘Rules, reasons and routines: Legislative roles in parliamentary democracies’, The Journal of Legislative Studies, 3(1), 155-174.

Tavits, M. (2009). ‘The making of mavericks: Local loyalties and party defection’, Comparative Political Studies, 42(6), 793-815.

Weeks, L. (2009), We Don’t Like (to) Party. A Typology of Independents in Irish Political Life, 1922–2007. Irish Political Studies, 24(1), 1–27.

Weeks, L. (2010). ‘Minor parties: a schema for analysis’, Irish Political Studies, 25(4), 481-501.

Weeks, L. (2011). ‘Rage Against the Machine: Who is the Independent Voter?’, Irish Political Studies, 26(1), 19-43.

Weeks, L. (2014). ‘Crashing the party. Does STV help independents?’. Party Politics, 20(4), 604-616.