ECPR paper for the 1999 workshop in Mannheim 990326-31
Candidate selection to the European Parliament:
A comparative study of Ireland, the Netherlands and Sweden
By
Magnus Blomgren
Department of Political Science
Umeåuniversitet
901 87 Umeå
Sweden
Tel: +046 90 786 95 42
Fax: +046 90 786 66 81
E-mail: magnus.blomgren@pol.umu.se
2
Introduction
A fundamental, but commonly underestimated aspect of democracy is the selection of candidates
that run for political office. In western democracy, one basic notion is that we elect people who
will represent us in different institutions. They shall embody our opinion, or to use Hanna Pitkins
words, they shall do ”substantive acting for others”1
. In the European concept of democracy,
party organisations play a crucial role in this process2
. The parties are the main apparatus to
conduct this selection of representatives and present these to the voters. In a sense, however,
this is an idealistic description of democracy. Parties’ motives are not solely to give the voters
opportunity to select a representative. Party leaders also have their own reason for monitoring
who is selected. On the one hand, they have to present candidates, which are possible successes
in terms of electoral support. On the other hand, they have to control that the homogeneity of
the party prevails. The process of selecting candidates is therefore not only a way to give the
voters the “best” candidate, it is also an arena for intra-party conflicts. The rules surrounding
this process, i.e. who decides who will be presented to the voters, is therefore important in
representative democracy. The notion of who and what representative’s represent is influenced
by who it is that selects them. Therefore, the analysis of the intra party selection process is vital
to understand representative democracy.
My focus in this paper is to analyse the way parties select their candidates to the European
Parliament (EP). I will compare the social democratic parties, leading conservative parties and
the green parties in the Ireland, the Netherlands and Sweden. In this paper, my ambition is not to
explain why the parties have chosen different processes of selection, but rather to discuss how
national parties organise their selection of candidates to the European Union arena. Starting
from this level of ambition the main purpose of this paper is to survey differences and similarities
in procedural rules than can be important for a further study of representation involving a
comparison between MP’s at the national level and the representatives at the European Union
level, i.e. MEP’s. The concrete research question is three fold. First, are there differences in how
the parties from different party traditions and countries select their representatives to the EP?
Second, are there differences in how the parties organise their selection of candidates to the
national legislative body and to the EP? Thirdly, too what extent are the party leadership
involved in the process of selecting candidates to the EP?
This study stems from two sources. In part it connects to the discussion about the political
legitimacy of the European Union. One of the main problems with the European Union is the
lack of a party system. Thus far, the integration of political parties and actions taken by political
parties on the EU arena, has exclusive been an elite-level phenomenon3
. This has major
implications to the concept of representation on EU level. This is often labelled the “democratic
deficit” of the EU and the current situation has made some writers to call for a new recruitment
system of European politicians 4
. In part the research question is emanating from the current
1
Pitkin, H.F. (1972), The Concept of Representation, University of California Press, Berkeley, 209ff
2
Gallagher, M. and Marsh, M. (ed.) (1988), Candidate Selection in Comparative Perspective – The Secret
Garden of Politics, Sage Publications Ltd, London, pp 2
3
Johansson, K-M. (1996), The Nature of Political Parties in the European Union, in Kourvetaris, G. A. and
Moschonas, A. The Impact of European Integration, Preager Publishers, Westpoint, pp 201 ff
4
Bryde, T. (1998), Party Groups in the European Parliament and the Changing Recruitment Patterns of MEP’s,
in Bell, D.S. and Lord. C. (1998) Transnational Parties in the European Union, Ashgate, Hampshire, pp 201ff
3
discussion about the new type of party organisations that are developing in Western European
democracies5
.
Theory and method
Representation in the European Union
In recent years, we have seen a huge amount of research regarding the new political system that
is developing in Europe. Much of this is of course both interesting and important work. At the
same time, something is missing. Most of this work misses the relationship between the developing
European political arena and the nation-state institutions and political parties. In other
words, a number of studies on the EU focus either on individual actors, such as a party or an
institution, or the ”horizontal” relationship among EU-actors. Often a vertical relationship is
simply assumed but not explicitly stated. The common assumption is that the national party
organisations will develop into European party structure. The result is that scholars often are
preoccupied with analyses of the developing European arena, but neglect to analyse their own
presumption that the process is guided by a vertical relationship. However, we should not only
consider the horizontal perspective and miss the vertical. My work is focused on one of these
vertical relationships, namely the internal party arena and the Members of European Parliaments
(MEP). Underlying this is my interest in the concept of representation.
The EU is commonly regarded as a system of negotiations. This is mainly due to the character of
the union, an organisation designed for international co-operation, but with a growing federal
capacity. The party organisations are represented both through their governments and through
representatives in the EP. This puts the party organisations in a new situation. On the one hand
they should enforce party politics in the EP, on the other hand they should be “loyal” to the
government policy in the Council of Ministers. We have two different representative ideals
present in the EU, political/ideological and national interests.
Given the new situation with an increasing importance of the EU arena, looking at the selection
process of candidates to the EP, at least two different types of interpretations are possible. (1) It
is possible that this new arena, with a low level of scrutiny and control by the national party
organisations, creates an incentive for the national executive branch to control the selection of
representatives to the EU arena. It is important that the party preserve its homogeneity inside
and outside the national boarders. In contrast, (2) it is possible that the attention from the
national executive branch, regarding selection of candidates to EP, will continue to be limited,
given the relative lack of political power of the EP.
In the study of candidate selection and political representation, the distribution of power within
the party organisations is of vital importance. In this sense, a hierarchical perspective is natural.
The leader is in need of support, but also wants freedom of action6
. The unequal nature of the
exchange can mean that the followers obtain only organisational incentives whereas the leaders
obtain both support and freedom of action. It is also likely that the greater the freedom of action
won by the leaders in a vertical power game (inside an organisation), the stronger they are in the
horizontal power games (outside the organisation). In this perspective the interaction between
5
Katz, R.S. and Mair, P. (1995), Changing Models of Party Organization and Party Democracy, Party Politics,
Vol 1, pp 5-28
6
This discussion is influenced by Panebianco, A. (1988), Political Parties: Organization and Power, Cambridge
University Press, Cambridge, 22ff
4
the members and the leaders is crucial for an understanding of the selection of candidates in
parties.
In political science, there have been few attempts to analyse the process of selecting candidates.
One reason is that party organisations have been reluctant to reveal their internal rules and
practices regarding these types of processes7
. There have also only been a few studies that have
a comparative ambition. The reason is of course that parties, parliamentary systems, electoral
systems, traditions etc, varies in different countries. It is hard to compare internal party rules,
which operates in different environments. Nonetheless, there have been a few theoretical
attempts to construct categories, which can be used for a comparative study.
The work of Gallagher and Marsh analyses several western democracies. In trying to explain
variations between different parties and countries they mainly use five different factors8
. First,
the legal provisions of the country, i.e. legislation about who is acceptable as a candidate or
rules demanding primaries etc, quite naturally influences the way parties conduct their selection
of candidates. Secondly, the political structure of the country. For example, a federal system
tends to influence the way parties are organised. Thirdly, the electoral system influences the way
parties select their candidates. The smaller the role of the voter in deciding which of the
candidates is elected the greater the power of the parties’ internal hierarchies. When electoral
systems provide for preferential voting, allowing electors to choose the individuals they wish to
represent them, local agencies can argue that the party ticket must be assembled carefully, with
sensitivity to the voters’ wishes. The local organisation is in a position to know just what ticket
will be of most appeal to voters in the constituency. Gallagher and Marsh’ last explanatory
factor is the nature of the party. Ideological parties will be more rigorous in the demands they
make of candidates, while pragmatic parties’, being entirely election oriented, will be more
tolerant of any candidate who seems likely to be popular at the election.
In the final chapter, Gallagher concludes that all of these factors have to be regarded as
explanatory. Legal provisions and the political structure of the country certainly influence the
way parties organise their selection process. For example, federal countries tend to have a more
decentralised structure, than parties in unitary states. However, it is easy to find counterexamples.
Some parties in federal states have a highly centralised selection process (f ex
Austria), and the opposite, some parties in unitary states has a highly decentralised structure (f
ex Britain). Furthermore, not even the electoral systems have a deterministic influence on the
structure of the parties, as their hypothesis implied. Gallagher can not find any evidence that, for
example, PR systems foster more centralised structures of selecting candidates. There is no sign
of a relationship between centralisation and the degree of voter choice permitted by the electoral
system. However, Gallagher finds some support for the assumption that political culture tend to
influence how parties organise this process. In some cultures, it is accepted that parties have the
total control over the selection process, while it is totally unacceptable in others, i.e. USA.
Lastly, the nature of the party ideology does not find strong support in the Gallagher and Marsh
research. For example, there is no tendency that socialist parties are more centralised than
bourgeois parties. All of these five explanatory factors have relevance for the outcome, but nonof
them really explains the whole picture. Rather, each has a bearing on the precise nature of the
selection process in a party. “To some extent, each party must be looked on as being sui generis.
7
Skare, A (1996) Kandidatutvelging – mer enn riktig kjønn fra rett sted, Tidskrift for samfundsforskning, Vol 37,
no 3, pp 331
8
Gallagher, M. and Marsh, M. (ed.) (1988), aa, pp 8-12
5
Parties do have some autonomy; their behaviour is not determined absolutely by their
environment”9
.
How to analyse the process of candidate selection
The process of selecting candidates is a complicated story. Parties use different systems and
works within different environments. To be able to make a fruitful comparison you have to
identify the steps in the process, identify the actors involved and create categories that captures
which actors operate at a certain step of the process. I concentrate on the formal rules and
procedures of the parties. This is a problematic strategy, because much of the substantive
selection of candidates is performed behind closed doors in a complicated system of interaction
between different actors10
. However, analysing the formal rules gives us a broad notion on how
parties organise their selection of candidates and information on the formal limits for different
actors in the process.
For analytical reasons we divide the selection of candidates into three steps. Appendix 1 is an
attempt to describe the process in a scheme. This is a simplification of a complicated process.
The purpose is to be able to create categories in order to make a fruitful comparison. All of the
three procedural steps can involve a different set of actors. Let me briefly explain the general
step-wise logic of Appendix 1.
First, we have a nomination procedure, where different candidates are proposed. Voters can be
allowed to nominate candidates (described in Appendix 1 as free nomination). This is probably
very seldom found in Western Europe. Most often there are some kind of involvement of the
party members. Either you have an intra party nomination or a subset of members nominates the
candidates, but you can also have a selection committee that has the right to nominate the
candidates. The fifth example is when the national executive themselves nominates all
candidates, but this category is probably unrealistic.
Second, we have the selection process. Someone has to select among the nominees and the
selection procedure is perhaps the most important of the three. It is the selection that I will try to
identify, and then analyse the procedures before and after the selection, i.e. the nomination and
decision procedure. In the selection procedure, the voters can be involved through inter party
primaries. Although, this system is found in other western democracies, it is unusual in the
European context11
. Members are involved through intra party primaries or a subset of members
that select among the nominees. Also in this procedure, you can have a selection committee or
the national executive that conducts the selection. The selection procedure, independent of who
make the selection, can be binding or non-binding. In the case it is binding, the selector is the
actual decision-maker. However, we can also find a variety of systems with non-binding
decisions. For example, were the selector selects a group of candidates, but does not decide the
order between them, it might be possible for some other organisational body to make additional
nominations, but not exclude the candidates chosen by the initial selectors.
Third, we have the actual decision. Who will be the candidate put forward to the voters? The
difference between the selection procedure and the decision procedure is important.
Hypothetically one set of actors (f ex regional branches) could select among proposed
9
Gallagher, M. and Marsh, M. (ed.) (1988), aa, pp 265
10
Gallagher, M. and Marsh, M. (ed.) (1988), aa, pp 5, for a more extensive discussion.
11
Gallagher, M. and Marsh, M. (ed.) (1988), aa, pp 236
6
candidates, but in the end another actor can decide which ones of those who actually will be the
candidates. This happens, for example, when the executive branch has a mandatory right to
confirm a candidate or a set of candidates.
Between these three procedures we often find some kind of intermediate organisational body,
that is designated to prepare for the selection procedure. This can include the right to organise
the next step in the process and/or to make a pre-selection of candidates. This organisational
body can be of importance but sometimes it is negligible. Hypothetically, they can influence
which one of the nominated candidates that the selectors or decision-makers are able to choose
from, for example by downsizing the list of nominated candidates.
The selected parties?
I have selected three countries for this study; Ireland, the Netherlands and Sweden. Within these
countries I have selected three parties: social democratic parties; Labour Party (LP), Partij van
de Arbeid (PvdA) and Socialdemokratiska Arbetarpartiet (SAP), the right wing parties; Fianna
Fáil (FF), Christen Democratisch Appel (CDA), and Moderata Samlingspartiet (MSP) and the
three smaller green parties; Green Party (GP), Groen Links (GL) and Miljöpartiet de Gröna
(MdG)12
. As for the comparative reason I have chosen countries that differ with respect to their
electoral system. The parties, within these countries, differ with respect to size and ideological
heritage. By choosing these three countries and parties, I have wanted to ensure variation while
at the same time keeping the sample at a size in which an in depth analysis of rules and
procedures is still possible.
A fundamental issue, regarding this analysis, is of course how the electoral system is
constructed. Several countries have a different number of constituencies for the national
assembly and the EP. For example, if a country have a PR system with one single constituency in
the EP election, but several constituencies at the national elections, this will probably influence
the way parties organise their selection process. In the 1994 European Election Study the
candidates were asked to estimate the power of different party bodies in their selection process.
When Pippa Norris analyses the result she concludes that countries characterised by elections in
which the whole country forms one consistency have highly centralised recruitment process.
Ireland and Britain had the most localised recruitment, and these countries use single
transferable vote (STV) and simple plurality system13
. This can explain the differences between
two sets of selection procedures within a party, but not necessarily explain differences between
different type of parties. In table 1, I present important characteristics of the electoral systems in
the countries that I study.
The research on the relationship between the electorate, candidates and party organisations is
rather limited. It is common to forcus on the outcome of elections, including variables like
proportionality, government strength and party system. However, I want to focus on the intra
party relationship. Douglas Rae analyses the consequences of different electoral system on the
inter party relationship14
. He categorises different electoral systems by defining different types of
ballots used in the electoral systems. Since, Rae is focusing on the relationship between parties,
12
Henceforth I will use the abbreviations mentioned in parenthesis
13
Norris, P. (1996) Legislative Recruitment, in LeDuc, L., Niemi, R. And Norris, P. (1996) Comparing
Democracies, Sage Publications Ltd, London, pp 199
14
Rae, D.W., (1975), The Political Consequences of Electoral Laws, Yale University Press Ltd, London
7
his categories are useful also for this study. The first category is whether the system uses
categorical or ordinal ballots.
Categorical ballots ask the voter to decide which one of the parties he/she prefers. The ballot
forces him/her to say that he/she prefers one party to all others. In some system you have a
simple candidate ballot, were you can vote for one candidate, but not different candidates from
different parties. In other cases you have simple party–list ballots, were you vote for a list of
candidates put forward to you by the party organisation. Rae does not differentiate between
those categorical ballots with fixed lists and those ballots were the voter might change the order
of the candidates within one party. This is not important in Rae’s study of inter party
relationship, but it is important when we analyse the intra party relationship. In table 1, I include
this in the column for intra party preference voting.
In electoral systems that Rae categorises as ordinal ballots, the voter is able to express a more
complex preference by ranking candidates from different parties. An example of ordinal ballots is
the alternative vote or STV, were the voter is asked to rank his preferences among the
candidates of different parties. With the ordinal ballots, the voter always have the possibility to
influence on the order of the candidates within one party or several.
The three selected countries have electoral systems with proportional representation, as is
described in table 1. In this sense, these cases are similar to each other, but in several other
aspects, their electoral systems differ. The most important aspect is if the system is characterised
as candidate-oriented or party-oriented. That is, if the electorate have the possibility to choose
between different candidates in the same party or split their votes between several parties
(candidates), the possibility of the party to control the representative is weaker than if the
electorate only chose candidates on fixed lists15
. These two different characteristics have an impact
on the way that representation works and the role of the party organisation.
Table 1 Electoral system characteristics in Ireland, the Netherlands and Sweden
National elections EP elections
Country Electoral
system
Ballot type Intra
party
pref
voting
No of
con-
stituen
ciesa
Elector
al sys-
tem
Ballot type Intra
party
pref
voting
No of
con-
stituen
cies
Ireland PR Ordinal:
candidate ballot
Yes 41 PR Ordinal:
candidate ballot
Yes 4
The Netherlands PR Categorical:
party list ballot
Yes 1 PR Categorical:
party list ballot
Yes 1
Sweden PR Categorical:
party list ballot
Yes 28+1 PR Categorical:
party list ballot
Yes 1
Note: a
Two or more numbers indicates multiple tiers.
Sources: Rae, D.W., (1975), The Political Consequnces of Electoral Laws, Yale University Press Ltd,
London. LeDuc, L., Niemi R.G. and Norris, P. (1996) Comparing Democracies, Sage Publications Ltd,
London. Cox, G.W. (1997), Making Votes Count, Cambridge University Press, Cambridge.
Generaldirektoratet för Forskning, Lagarna För Val Till Europaparlamentet, Politiska Serien No.13,
Europaparlamentet, 1997
15
Katz, R.S. (1980), A Theory of Parties and Electoral Systems, John Hopkins University Press, Baltimore, pp
116
8
Ireland has a bicameral system, were the Upper House is elected through a complex system of
electoral colleges, panels of interest groups and by the PM16
. The Lower House is elected
through proportional representation via the single transferable vote (STV)17
. It is a candidateoriented
system operating within 41 multi–member constituencies, in which voters are expected
to rank the candidates in order of preferences. In the EP election the number of constituencies is
reduced to four. Candidates from the same party are therefore often in open competition with
one another18
. In this system voter’s vote for candidates, not parties19
. The Netherlands has a
bicameral system, where the first chamber is elected through the provinces while the second
chamber is directly elected. The Dutch electoral system is characterised as extremely
proportional with the whole country as one constituency in ordinary national elections as well as
in EP elections. Netherlands does not have a formal threshold, which of course increases the
proportionality. The electoral system is one with a single preferential vote. Though the parties
are able to present different lists for different sub–districts, which gives a hypothetical
opportunity for the electorate to change the list–order, this has virtually no impact on the list
order presented by the party. The Dutch system is therefore to be characterised party oriented
and not candidate oriented20
. The Swedish electoral system is regarded as highly proportional. It
is a two–tier system where the lower tier consists of 28 constituents covering the country in
ordinary national elections, which between them return 310 deputies. The remaining 39 seats are
held back for allocation at the second tier with the purpose to ensure that the total number of
seats received by each party comes as close to its proportional share as possible. In the EP
election Sweden is one constituency. In both national and EP elections there is a threshold at 4%
of the votes, which discriminates against small parties. The list of candidates is put together by
the parties. In the 1998 election new electoral rules was used for the first time. Before, there was
a hypothetical opportunity for the electorate to change the list order, but in practical terms this
was nearly impossible. In the 1988 election it was added a possibility for the voter to cast a
preferential vote for one candidate on one party list. A candidate needs 8% of the votes for the
party in the electoral district in order to alter the ranking of the party organisation. Despite the
new rules, I categorise Sweden as a party oriented electoral system.
In the selected countries there are approximately 25 parties represented in the national parliaments.
It is not possible to study all of these and as mentioned above, I therefore have to select
some of these parties. The parties shall differ in the respect of size of the party and placement on
policy dimension.
First, we examine the different size of the parties. In order to give some perspective to this
question I compare the selected parties, with their counterparts in other European countries. In
table 2, I describe the level of organisation for left wing parties (Social Democratic Parties),
16
For a more extended description and discussion see Chubb, B. (1992), The Government and Politics of Ireland,
Longman Group Limited, London, pp 197
17
For an extensive and excellent description of the STV system see, Busteed, M.A. (1990), Voting Behaviour in
the Republic of Ireland, Clarendon Press, Oxford, pp 47-73
18
Farrell, D.M. (1994), Ireland: Centralization, Professionalization and Competetive Pressures, in Katz, R.S. and
Mair, P (ed), How Parties Organize, Sage Publications Ltd, London, pp 220
19
Gallagher M, Laver, M and Mair P (1995), Representative Government in Modern Europe, McGraw-Hill Inc,
New York, pp 287
20
Koole, R. (1994), Ireland: Centralization, Professionalization and Competetive Pressures, in Katz, R.S. and
Mair, P (ed), How Parties Organize, Sage Publications Ltd, London, pp 282f
9
right wing parties (mainly Conservative or Christ Democratic Parties) and the green parties, in
some of the EU countries21
.
Table 2 Level of organisation in some EU countries 1989
(Members divided with the electoral support in %)
Country Left-wing party Right-wing party Green party
Austria SPÖ 29,5 ÖVP 24,8 GA 0,9
Belgium PS/BSa
13,6 PSC/CVPa
10,0 ECOLO 0,8
Denmark SD 9,9 KRF 6,3 n/a n/a
Finland SDP 12,3 KOK 10,6 n/d n/d
Germany SKP 6,6 CDU 5,1 DG 1,2
Ireland LAB 4,3 FF 12,0b
GP 4,8
Netherlands PvdA 3,4 CDA 4,0 GL n/d
Sweden SD 9,7c
M 8,2 MP 3,0
U. Kingdom LAB 2,9 CON 5,5 n/a n/a
Note: a
This is calculated as the mean result of PS and BS respectively PSC and CVP.
b
Thus is calculated on the basis of information from Chubb, B. (1992), The Government and Politics of Ireland,
Longman Group Limited, London, pp 79, 105 and 136. The membership figure is an approximation done by the
party head office and is probably an overestimate according to the Chubb.
c
A system with corporate members was in practice in 1989. If we include the corporate members the
organisational level for the party should be 46,6% in 1989. However, this was abbolisched in 1991 and thereafter
the level of organisation has declined. Between 1990 and 1992 the party lost 775.000 members and that would
mean 9,7% in organisation level based on the 1988 election. Bäck, M. and Möller, T. (1997), Partier och
Organisationer, Norstedts Juridik AB, Stockholm, pp 112.
Source: If no other reference is mentioned, the source is Katz, R.S.and Mair, P. (1992), Party Organizations: a
datahandbook on party organizations in western democracies, 1960–1990, Sage, London
The first conclusion is that the left wing parties organise on average more of its supporters, than
the right wing parties. However, the difference is not that huge (10,2% for the left wing and
9.6% for the right wing22
). The other general conclusion, although this is based on a small
number of cases, is that the green parties have much more votes, than members. This indicates
that there is a fundamental difference between these two types of parties. The new green parties
receive many votes compared to the number of members, i.e. in this perspective; they are more
of plain parliamentarian parties compared to the others.
If we then turn to the different countries, we see that the Nordic countries, Belgium and Austria
seem to have a stronger organisation level compered with the others. This is of course a
problematic discussion, given that I only present figures for three types of parties. However, this
conclusion is consistent with other studies in this field23
. The Nordic countries, Belgium and
Ireland all have organisation levels 7–11,5% for the presented parties. Austria represents the
extreme case with an organisation level at 18,4 % for the presented parties. The party organisations
major role in Austria is described as “…society itself has been colonised by party to an
extent in which almost all societal organisations (from the automobile associations to the major
conservationist organizations) bear a party label or have close links with one of the two major
21
Membership figures are a problematic theme in political science. Often parties keep a poor record of their
members and they are not that easy to compare. The lack of information makes it difficult or perhaps
impossible, to include all the countries and parties in the EU. Although these shortcomings, it is interesting to
notice some systematic patterns among the parties that are described.
22
In this calculation I have adjusted the figures for the Swedish SD to 1992 level.
23
Katz, R.S. and Mair, P. (1994), How Parties Organize, Sage Publication Ltd, London, pp 5
10
parties”24
. Germany, Netherlands and United Kingdom are the cases with the comparatively
lowest organisation level, between 3,7–4,6% for the presented parties. This means, for our
selected countries and parties, that the organisational level is higher in Sweden and Ireland than
in the Netherlands. This conclusion is of course problematic, due to the highly unreliable figure
especially for FF25
and the fact that we do not have any membership figures for the GL.
The left-right dimension is not that obvious in the Irish case, as in the other selected countries26
.
The origin of the Irish party system is partly the struggle for independence. As I have mentioned
before, FF is the major government party in Ireland. It is also the right wing party, which has
been most successful in elections in the selected countries. LP have been in power in some post
war government, but in terms of electoral success and participation in government it does not
compare with the other to social democratic parties in Sweden and the Netherlands. LP is
relatively smaller than the other two. Today, FF is in a government coalition with Progressive
Democrats.
CDA have a centre position in the Netherlands. CDA is a result of a merge in 1977 between the
relatively large Crist Democratic Parties in the Netherlands27
. Due to its intermediate position,
CDA and its predecessors have been in every Dutch government during the post war period.
PvdA is the Dutch Social Democratic Party and have been in several governments with CDA or
some of its predecessors. It is relatively large compered to LP in Ireland, but smaller than SAP.
At present time PvdA form a coalition government with Democraten 66 and Volkspartij voor
Vrijheid en Democratie.
In Sweden, the major governmental party is SAP. It has been supported by about 45% of the
electorate during the post war period, but has had a declining tendency from mid 70’s. The right
wing counterpart, MSP, has increased their support during the 80’s and is now the dominant
opposition party in Swedish politics. After the most recent election in 1998 the SAP again
formed a single party minority government.
The Green parties are the smallest party in each country. They are also comparatively new
organisations, which is interesting in a comparative perspective. Non-of the studied green parties
have ever been in a government position.
Empirical study
We now turn to the empirical study. This section is divided into two sub-sections. I start every
country section with a general description of the normal procedure for national elections. After
this, I examine the selection procedure for the EP election, party by party. We conclude the
24
Müller, W.C. (1994) The Development of Austrian Party Organizations in the Post-war Period, in Katz, R.S.
and Mair, P. (1994), How Parties Organize, Sage Publication Ltd, London, pp 52
25
The estimation from the FF head quarters is 89.000 members in 1989. At the same time the general secretary
of the party claimed that the party had 30.000 members in 1992. This was presented at the 60th
party
conference, according to Allen, K. (1997), Fianna Fail and Irish Labour – 1926 to the present, Pluto Press,
London, pp 184. This seems to be an unrealistic decline in membership figures. In Irish Political Studies, vol.
8, 1993, pp 202 the estimate from the FF head quarter is 75.000. In Irish Political Studies, vol. 9, 1994, pp
231, the report is still 75.000, but with a footnote that the party claims to have 30.000 active members and the
rest should be consider as “associated members”.
26
Budge, I. (1996, Great Britain and Ireland – Variations on Dominant Party Government, in Colomer, J.M.
(1996) Political Institutions in Europe, Routledge, London, pp 22
27
The merge was between katholieke Volkspartij (KVP), Anti-Revolutionaire Partij (ARP) and ChristelijkHistorische
Unie (CHU)
11
country sections with a comparison of the parties within the country. After this, I compare the
countries and the different types of party organisations according to the analytical scheme that is
proposed in this study. (The coding of the selection process for the EP elections in different
parties is summarise in table 3 on page 21.)
Sweden
Sweden uses list system and multi-member constituencies. The party organisations are normally
divided in three levels, the local branch, regional level and national level. The local level and the
regional level are in most case synonym to the municipalities and regional public administrative
level. At ordinary national elections, the selection of candidates is a fairly decentralised process
in the Swedish party organisations. Generally, parties put forward lists in every one of the 28
constituencies and the decisions are often legitimised through regional electoral conferences.
The use of non-binding intra party primaries has become more common in the Swedish parties;
more and more parties use it on a regular basis. However, MSP is the only party that has used
intra party primaries regularly for many years. Intra party primaries are never fully binding in the
Swedish parties. You often have an electoral conference, a regionally based selection committee
or some other kind of decision making body that makes the final decision about the list.
The selection process of EP candidates in SAP (Socialdemokratiska Arbetar Partiet) begins with
intra party nominations at the regional level28
. All members have the right to nominate and the
nominations are sent to the party apparatus on the regional level. The board on the regional level
selects some of the nominated candidates and describes them as the candidates that are to
represent the regional level. All the other nominated candidates are mentioned as “surplus”
(övriga) nominations. The board on the regional level can be described as an intermediate
administrative body in my categorisation and it is able to influence the nomination process.
The nominations are then sent to the national executive committee, which put together a list of
candidates and forwards this list to the national board. The national board is able to make
changes in the proposal and they are obliged according to the statutes to consider an equal
gender allocation, but also age and regional representation. The party has also decided that the
candidates have to accept the position of the party concerning the future direction of the EU29
.
This is a result of the split in the party on the issue of membership in the EU. According to EPmembers
of SAP, the present split in the group has reduced the ability to perform an active EUpolicy.
This is something that some highly placed party representatives wants to resolve with the
1999 EP election, i.e. they want to avoid presenting candidates with a negative view on the
membership issue30
. According to the statutes, the national party board is the one that decides
the final disposition of the list. However, the national board has delegated this capacity to a
newly created forum, the body of regional representation (Förtroenderådet). This is an
organisational body created to scrutinise the national political apparatus in the party. Formally
Förtroenderådet is not a level of decision making in the party, but in the process of selecting
candidates for the EP election, the board has decided that Förtroenderådet makes the actual
decision in the name of the board. Consequently, the Förtroendrådet is able to change in the
28
The material used for the Swedish Social Democratic Party is the party statutes from 1997 party congress and
an interview with Mr. Stefan Svensson (national organisational chief) 981118 and 990119 and Sven
Hulterström (Parliamentary party group leader) 990122
29
Interview with S. Svensson 980119
30
See f ex Svenska Dagbladet, 990131, Schori hotar Theorins EU-plats and Dagens Nyheter 990110, Kritiker bör
inte ställa upp i EU-valet.
12
proposal of the board and the board is not able to change the decision of the Förtroendrådet.
They can change the order of the candidates and also include candidates nominated by the
regional level.
In MdG (Miljöpartiet de Gröna)31
the process of selecting candidates to the EP election starts
with an intra party nomination procedure. All members are entitled to nominate candidates. The
nominations are sent to the national selection committee, that organises the next step in the
process. The selection committee is not able to influence the process at this stage. All the
nominated candidates are presented in the internal paper in alphabetic order.
The next step is an intra party primary. All members have the right to vote for a candidate and it
is the selection committee that administers the primary. This is the first time that MdG organises
a primary. The selection committee considers the result of the primary and decides on a list of
candidates in a fixed order. The selection committee is obliged to consider an equal gender
allocation.
The proposal is presented to an “extra congress”, which makes the final decision. This extra
congress can alter the proposal from the selection. The party board has no right to change the
decision of the extra congress. One important note is that the nomination procedure does not
end with the primaries. As a member you have the right to nominate new candidates until a fixed
time at the extra congress. This could be important, given that there is an internal debate
concerning the party position on the long-term strategy in EU. Some members, among others
the chairman of the youth organisation, has criticised the party position in favour of leaving the
union32
. There has been a discussion within the party to form an alternative list that contains
candidates with a more positive attitude towards the union. This strategy was never put to a test,
but the proposal from the selection committee deviates from the result of the intra party primary
and one reason is that some candidates had a different attitude towards the EU, according to the
chairman of the selection committee33
.
MdG has a gender quota rule in their statutes (40% minimum for any gender). However, the
candidate selection procedure does not give any organisational body the right to alter the list, as
an effect of this rule, after the decision is taken on the extra congress, i.e. it is sovereign, even if
it violates its own gender quota rule.
In MSP (Moderata Samlingspartiet)34
all members are entitled to nominate candidates for the EP
election. After this an intermediate procedure takes place. All nominated candidates are
presented to the party board, the regional representatives at the annual national congress
(partistämmoombud), and its substitutes. Members of this group votes individually for 5-10
candidates. Based on this preliminary procedure, the party board picks out 60 candidates. The
party board is entitled to add candidates to the list up to the limit of 60 candidates.
31
The material used for the Swedish Green Party is the party statutes from 1992 party congress (revised at 1993
and 1994 party congress) and an interview with Kjell Dahlström (General Secretary) 990120 and Marianne
Samuelsson (Parliamentary party group leader) 990121
32
Dagens Nyheter, 981215, EU-Fronten i mp rämnar.
33
Dagens Nyheter, 991229, EU-positiva petas från mp:s EU-vallista.
34
The material used for the Swedish Conservative Party is the party statutes and an interview with Cleas Löfgren
(party board) 981102 and Gunnar Hökmark (General Secretary) 990203
13
The list of candidates is then put forward to the members in alphabetic order and an intra party
primary procedure takes place. All members are entitled to vote and they shall rank five
candidates. The vote is sent to the national party administration. The party board, on the basis of
the result in the primaries, then determines the list of candidates. The board is entitled to add
candidates after the primaries and they are not obliged to follow the result of the votes. In the
statutes it is mentioned that the party board should consult with the regional chairs before a
decision is taken.
The three Swedish parties in the study all show similarities as well as differences. They all use
intra party nominations, but they differ in the selection procedure and they use different kind of
decision procedures. In ordinary national elections, the process is more decentralised, due to the
fact that the electorate is divided into 28 constituencies. At ordinary national elections the
selection of candidates takes place at the regional party level. In the EP election the entire
country is one constituency. The interviewed party representatives mention this as the main
reason for the more centralised procedure in the EP election.
At the same time all the parties have tried to legitimise the selection of candidates, as in SAP,
through a body consisting of regional representative (Förtroenderådet), or through primaries as
in MSP and MdG. In the SAP it is becoming more common to use different primary procedures
in the regional party branches before a national ordinary elections, according to Svensson.
However, in the EP election this procedure is not used, because the party does not have a
centrally administered membership matrix. Due to this an intra party primary would be too
complicated, according to Mr. Svensson, the national organisational chief. As mentioned above,
MdG arranges a primary for the first time. However, the result was a disappointment. A very
small number of members used the possibility to vote for a candidate. According to the general
secretary Mr. Dahlström, approximately 10% of the members voted. MSP usually organises
intra party primaries on constituency level in ordinary national elections. The party organisation
and the members are used to this kind of procedures. The primary for the upcoming EP elections
resulted in a turn out of 20%, which is the highest turn out, compared to other Swedish parties.
In MSP and the SAP the executive committee and the party board has an obvious influence in
the decision process. In SAP, they prepare the proposal for the representative body
(Förtroenderådet), and in MSP the party board is entitled to alter the list after the primary is
ended. The media reports from the nomination process also show that the national party
executives in both parties have influenced the process. The national executive committee of SAP
officially proposed one nominee as the top candidate, although the nomination process was not
over in the party35
. In a way this is natural according to the party rules, however, the media
reactions is that the top name is already selected and it is difficult for the Förtroenderåd to
change this picture. In other words, the party board has an obvious advantage in the selection
process.
In the MdG the influence of the national executive is less evident. The nomination procedure is
open for all members and all the nominations are put forward to the members in a primary.
However, the primary is not binding. The selection committee is able to alter the list after the
primary is taken place. A crucial question is the relationship between the executive power in the
party and the selection committee. The selection committee could serve as an emissary for those
in power in the party. Whether this is the case is of course hard to say, but in the media reports
35
Dagens Nyheter, 990109, Schori lanseras som kandidat till EU-valet.
14
from the nomination process, high party representatives have showed reluctance to comment on
the proposal from the selection committee and generally they have referred to the decision of the
extra congress36
.
Ireland
In Ireland the procedures for selecting candidates are very similar in all parties. The political
system in Ireland is often characterised by its locality and clientelism37
. This also has an effect on
how parties select their candidates. The electoral system is candidate based and the candidates
are normally selected by the party organisations within every constituency by selection
conventions. However, the national apparatus have in most cases some kind of influence on the
selection procedure. Often this is put into practice after a local or regional selection procedure
has taken place. The normal situation is that the executive committee of the party have the right
to add candidates or to veto candidates. The different national leading bodies of the parties in
the study have also added candidates in previous elections, although they all emphasise that the
selection of candidates is a local or regional responsibility. However, several studies of the Irish
party organisations suggest that the national party headquarters tend to exert an increased
influence on the selection of candidates38
.
In the LP (Labour Party) the selection for candidates to the EP begins in the local branches39
. It
is these who nominate the candidates and they are limited to one nominate one candidate each.
Usually a candidate get support from several branches, which in practice is a requirement in
order to be successful in the selection process. The process continues with intra party selection
conventions. There are four selection conventions in accordance with the four EP
constituencies. In an ordinary national election this number is 41. The local branches appoint
two delegates to the selection convention. In ordinary national elections every branch elects four
representatives to the selection conventions.
It is the party executive that decides how many candidates every convention is able to choose
and appoints the chairmen’s for the conventions. The executive committee administers the
nominations, but has no possibility to influence which candidates that are put forward to the
conventions. If a local branch nominates a candidate in a prescribed way, the candidate is in the
game until the convention says something different. On the convention a debate takes place and
the representatives argue for their candidate. In the end, the convention decides by proportional
representation with a secret ballot. However, the result of the convention is not binding. After
this procedure the national executive committee is able to change the result of the convention.
They can add candidates and hypothetically reject candidates. Although they have added
candidates in previous election, this is quite unusual, according to the general secretary, Mr.
Kavanagh. In the last EP election they actually added a candidate in one constituency and this
36
See for example Dagens Nyheter, 991222, Mp-krav på”ut ur EU” står fast.
37
See one description in Farrell, B. (1985), Ireland: From Friends and Neighbours to Clients and Partisans: Some
Dimensions of Parliamentary Representation under PR-STV, in Bogdanor, V. (1985), Representatives of the
People?, Aldershot, Glower, pp 237-264
38
See for example Mair, P. (1987), Party Organizations, Vote Management, and Candidate Selection: Toward
the Nationalization of Electoral Strategy in Ireland, in Penniman, H.R. and Farrell, B. (ed) (1987), Ireland at
the Polls 1981, 1982, and 1987, Duke University Press, Durham N.C. See also Marsh, M. (1995), The 1994
European Parliament Election in the Republic of Ireland, Irish Political Studies, Vol 10, pp 209-215
39
The material used for the Labour Party is the ”Standing orders for European Parliament selection convention”
and an interview with Mr. Ray Kavanagh (General secretary).
15
was strongly criticised within the party40
. It also turned out that this candidate did not succeed in
the election. This experience is still remembered in the party and probably has an effect on how
the national level decides to do in this EP-election41
. To reject a candidate is a formal option for
the executive committee, but would be very unwise, because of the opinion within the party. Mr.
Kavanagh is careful to point out that the selection of the candidates is a matter for the local
branches.
The LP has a rule that says that they shall have a gender quota at 25%, i.e. the minority can not
be smaller than this percentage. This is a rather modest ambition, according to Mr. Kavangh,
and it is not too problematic to fulfil through the ordinary procedures for selecting candidates. If
a problem should occur, it is the executive committee that is responsible to correct the result of
the selection procedure. The party also has a rule that the candidate can not run as a candidate
for an ordinary national election, while sitting in the EP. This is to avoid a double representation
in the party.
The rules for FF (Fianna Fáil) are similar to the LP42
. The nominations are put forward in the
local branch (cumainn) and they decides to nominate one candidate. To be regarded as a
candidate you need a certain amount of branches backing you up (50 in the Dublin area and 30
in the rest of the country). The nominations are administered by the FF head quarters, but the
national executive is not allowed to intervene in the process at this stage. The process continues
with the local branches appointing one representative to a selection convention. In ordinary
national elections the number of representatives from every local branch is three. The party
arranges four conventions, one per constituency. The executive committee decides how many
candidates the selection convention is allowed to select and they also select the chairman for the
convention.
The decision of the selection convention is not binding and the executive committee is able to
add candidates to the list of candidates. This has also been done in past elections, but according
to Mr. Mackin, the general secretary of FF, the party will not use this competence for this EP
election. According to Mr. Mackin, the reason for this back door option for the executive
committee, is that they want the opportunity to balance the ticket because of a disproportional
gender allocation or regional representation. The executive committee also has the opportunity
to veto a candidate, but this is rare and would not be popular in the party organisation.
There is a standard procedure that an elected MEP gets re-appointed as a candidate more or less
automatically at the conventions43
. This means that if you once have been an elected MEP you
occupy this position until you decide to step down. In practice, the selection at the conventions
is for the remaining seats of the constituency. This is probably the same in several parties in the
study, but it is not as explicit as in the case of FF.
40
The Labour leadership supported the candidacy of a young TV journalist, Orla Guerin, and when the
convention selected Bernie Malone, the national executive added her as a second candidate in the Dublin area.
See Marsh, M. The European Parliament Election in the Republic of Ireland, Irish Political Studies, Vol. 10,
1995, 209-215.
41
Se for example The Irish Times, 980624, Invitation to Cassells causes Labour row.
42
The material used for Fianna Fáil is an interview with Mr. Martin Mackin (general secretary) 990209 and with
Dr. Rory O’Hanlon (Leader of the Parliamentary Party and Deputy Chairman of the Dáil Éireann) 990211
43
See for example The Irish Times, 990208, Gallagher selected for Euro election.
16
According to Mr. Mackin, the main differences between an ordinary national election and an EP
election is not in the procedure of candidate selection, but in the type of campaign that will
follow. The constituencies are much larger in the EP election and the campaigns tend to be more
about the personality of the candidates, than of party politics.
The selection procedure in the GP (Green Party) differs in some respect compered to the other
two Irish parties44
. Every member is entitled to nominate a candidate, although, this is not
formally stated in the party constitution. It is the regional structure in the party that decides who
can nominate and arrange the selection procedure, either the local branches or direct
nominations from members. The common procedure is, though, that the nominations go direct
from the member too the regional office. The regional level is divided into the same areas as the
constituencies for the EP election. The selection convention that takes place is open for all
members. It is also possible for a member to give a postal vote for a candidate if he/she is unable
to attend. The party uses the same procedure in general national elections.
It is the regional “consult” that decides how many candidates every regional body can chose. It
is not decided in the central party apparatus. However, the party has never contemplated to put
forward more than one candidate in every constituency. The argument is of course, that too
many candidates can lead to splitting votes and loosing seats. Both the co-ordinator Mr.
Kearney and the member of the rules and procedure committee, Mr. Goodwillie, emphasise that
the party has a strong commitment to the decentralisation of the organisation. Therefore, the coordinating
commission (equivalent to an executive committee) is not able to interfere with the
procedure of selecting candidates. However, the co-ordinating commission is formally able to
veto a candidate, but this has never been done. According to Mr. Kearney, a hypothetical
example when this rule could be used is if a candidate, after he/she is selected, turns out to have
a history that would not be “appropriate”.
The party has a gender quota rule, which creates a bit of a problem in the selecting of candidates
to elections. The formulation in the statutes is more of an aspiration, rather than a requirement.
The reason is that the electoral system is constructed in a way, that in order to fulfil the gender
quota more rigorously, this deserves a more centralised candidate selection. This is a dilemma,
two goals collides, gender quota or a decentralised candidate selection procedure. According to
Mr. Goodvillie, hypothetically, if the party should put up two candidates they could impose a
gender quota. This means that the argument for a more centralised candidate selection could be
used if the party increased its support and won more candidates.
The differences among the Irish parties in their procedures for selecting candidates are relatively
minor. In fact, at the surface there are only two distinct differences. First, in how you get
nominated, and secondly the composition of the selection convention. In FF you need support of
a certain amount of local branches in order to be regarded as a candidate. In the LP every branch
have the right to nominate one candidate each (although in practice you need more support
when you go to the convention in order to get elected) and in the GP every member can
nominate whoever he or she wishes. However, it is hard to see that this difference should have a
fundamental impact on the way candidates get elected.
44
The material used for the Green Party is the statutes from 1997 and an interview with Mr. Phil Kearney (coordinator)
and Mr. John Goodwillie (rules and procedure committee) 990210
17
There is also a difference between the parties in composition of the selection conventions. FF
and LP differ in the number of candidates the local branches send, while the conventions are
open for every member in the GP either to attend at the convention or to use a postal vote. In
this respect the GP arrange a type of intra party primary, while the decision in the other parties
are conducted by a subset of members. However, it is important to note the difference in size of
the three parties. The branches in FF and LP are small and many. That every branch sends one or
two delegates to the selection conventions should probably be regarded as a fairly representative
decision making procedure. The GP is much smaller in this respect and it is probably not
necessary with a procedure of selecting representatives to the conventions. However, it is a
principally important difference and if the GP would grow and still maintain with their current
system, the differences could be of importance.
All of the three Irish parties emphasise that the selection of candidates is a matter for the local
branches and the constituencies. This is not at least due to the political culture in Ireland and to
the electoral system. The general secretary of FF, Mr. Mackin, expresses himself, “…because of
the proportional representation that we have and the clientelism that we have, you have to go
out there on the road”45
. As mentioned before, other studies have shown that the national party
headquarters tend to intervene more in the process today, than before. This study either verifies
or falsifies this conclusion. However, both the LP and FF explicitly states that they will not
formally interfere in the selection of candidates at the upcoming EP election.
The Netherlands
The electoral system in the Netherlands is not divided into constituencies. Nor is there an
attempt to try to avoid a fragmentation of the parliament with a formal threshold. From a
comparative perspective these are unusual institutional features. They effect how the parties
organise their selection of candidates. The Dutch parties are known to have a rather centralised
selection of candidates46
. The parties are commonly divided in three levels, local, regional and
national and it is not unusual that the parties present several lists in different regions. The point
is that the top candidates are the same on every list and the lower ranked candidates can have a
regional profile. Although, this is the case, most analysts seem to concur with the statement that
the “…ultimate decisions on the composition and ordering of each and every list lie inescapably
with the national party organisation”47
.
In CDA (Christien-Democtratic Appeal) the process begins at the top level with the creation of
a profile of which type of group and individuals that should represent the party in the
parliament48
. This concerns gender and regional allocation, types of skills and so on. At this time
there is also a screening of the candidates that are already present in the EP, if someone is
45
Interview with Mr. Mackin 990209.
46
See for example Gladdish, K. (1985), The Netherlands, in Bogdanor, V. (1985), Representatives of the
People?, Aldershol, Glover, pp 136 and Koole, R. and Leijenaar, M. (1988), The Netherlands: the
predominance of regionalism, in Gallagher, M. and Marsh, M. (ed.) (1988), Candidate Selection in
Comparative Perspective – The Secret Garden of Politics, Sage Publications Ltd, London, pp 201.
47
Gladdish, K., The Netherlands, in Bogdanor, V. (1985), Representatives of the People?, Aldershol, Glover, pp
136
48
The material used for CDA is the regulations for the EP candidate selection, ”Reglement voor de voorbereiding
van kandidaatstelling voor de verkiezing der leden van het Europese Parlament”, an interview with Mr. J.M.
Wiggers (head of the International Secretariat and one of the representatives in the selection committee for the
EP election) 990222, and an interview with Mr. MJM Verhajen (MP, spokesman on international issues and a
former MEP) 990223
18
leaving or if someone does not have the support from the party49
. It is the General board (party
board) that decides upon these issues. Thereafter, the process continuos with an appointment of
a selection committee by the General board. This selection committee consists of 6-7 people and
they represent different persons with experience of the work in the EP, representatives of the
party leadership and persons that represent the local branches. They also connect people as
advisors to this committee, as for example the chairs of the parliamentary group and the chair of
the foreign affairs committee.
This group starts the selection of candidates. People can “apply for the job”50
and the local
branches can nominate candidates. For a candidate to be accepted, the requirement is that five
local branches nominate him or her. The representatives already in the EP are automatically
included in the process. After this a first screening procedure takes place. The committee
interviews the candidates and the purpose is to select the candidates that are suitable for the
responsibility. Some will not get interviewed because they do not fit the profile. One
hypothetical example suggested by Mr Wigger (head of the International Secretariat and one of
the representatives in the selection committee for the EP election), is that a nominee with only a
primary school degree and that has reached the age of 65 is not likely to be interesting for the
committee. In the end, the committee makes up a list in alphabetic order and describes the
nominated candidates as very suitable, suitable and not so suitable. This list is forwarded to the
Daily board (equivalent to an executive committee) and they make up the final proposal for the
General board.
The General board takes into account that the list have to be balanced according to the profile
described earlier and that the list has a reasonable gender, regional and religious allocation. The
party has a 50% gender quota “rule” but this is hard to accomplish, according to Mr. Wiggers.
The party also tries to balance the representation between the different regions. In a system with
one constituency this type of monitoring is highly important. The General board also listens to
the group sitting in the EP, whether important people in one policy area are missing and so
forth.
The list is then sent to all the local branches and they vote on the proposal. This means that they
can go along with the proposal or change the order of the list. The result of the votes is sent to
the congress, that makes the final decision. The General board or the Daily board does not have
any formal right to influence the process at this stage. However, the normal situation is that the
voting of the local branches balance each other, i.e. if one local branch supports one candidate,
another local branch supports another, and the result is that the list made up in the General
board is accepted. According to Mr. Wedding, it is also very seldom that the congress changes
the list after the vote in the local branches, which means that the proposal from the General
board has a god chance of being accepted through the whole process.
The procedures are the same in an ordinary national election. However, there are a difference
concerning the number of candidates, which makes the process a little bit more difficult in the
EP election. In the EP election the required mandates to fill with candidates are more limited
than in a national election. According to Mr. Weddings this makes the balancing of the ticket a
little bit more difficult in the EP election. Everyone knows that it is more or less impossible to
49
The party also has an informal rule that a representative should not possess a mandate no longer than three
terms.
50
Interview with Mr. Wiggers 990222
19
fulfil every desire with the limited number of candidates that are to be selected. This makes the
game a little bit different in the selection procedures for the EP election.
The procedure in the PvdA (Partij van de Arbeid ) is similar to the one in CVA51
. The party
board begins with adopting a profile of the desired individual candidates and the group as a
whole. This is sent out to the local branches. At the same time a selection committee is
appointed. It is the selection committee that receives the nominations, individual applications
and the members of the committee makes their own scouting. The selection committee is
consisting of prominent senior members and representatives with international experience. The
party also advertises in newspapers for candidates. Other parties have tried to scandalise this
way of collecting nominations, however PvdA argue that it as a way to open up the nomination
procedure. The selection committee has to take into account the criteria of the profile. One of
these is an equal gender allocation, which the committee managed to achieve. The gender quota
rule is not obligatory according to the statutes, but it is an outspoken aspiration of the party. The
committee also monitors the regional representation and representation from the minority
groups. At the same time, there is a difference between the national election and the EP election.
According to Mr. Berkvens (assistant international secretary and a member of the PES
delegation), the different type of skills and expertise in areas such as social policy, agriculture
policy and so on, is more important in the EP election, than in the national election.
The selection committee makes a proposal where the different candidates are listed and attaches
its advice. This is put forward to the party board, which can make changes in the list, (this also
happened this time). The party board then forwards their proposal to the congress. This is a
difference between PvdA and CDA, in the last case the proposal from the party board is sent to
the local branches for a vote. In the case of PvdA, the proposal goes directly to the congress.
The congress is able to change in the proposal, but this is quite complicated. However, this has
been done in the past, although the last congress supported the proposal from the board.
There is no difference in how the candidates get selected to the national parliament and to the
EP. However, the party has changed their candidate selection procedures in the 90’s and the
procedures described are used for the first time in an EP election52
. In the past the party involved
the local branches in the process, they nominated candidates to the provincial level of the party
and the local branch had influence in the process when the list was completed. Today the
involvement of the local branches is minimal. The reason for the change was a disappointment
that the lists consisted of too many white middle aged male candidates. These were the ones that
got elected by the local branches. This was not representative for the image of the party and for
the Dutch population. By centralising the process, the monitoring of the result improved,
according to Mr. Berkvens. However, the new system is debated in the party. Many members
mean that the link between the party organisation and the candidates is weakening. Another
argument is that it is more difficult, in the new system, to get equal regional representation.
According to Mr. Berkvens, this change in the system was implemented when the party lost
many members and during a debate on the character of the party. He means that a change of the
focus of the party has occurred. Nowadays, they concentrate more on the electorate than on
members in the local branches. The new way of selecting candidates is one result of these
51
The material used for PvdA is the party statutes from 1998 and an interview with Mr. Arjen Berkvens
(Assistant international secretary and a member of the PES-delegation) 990224 and an interview with Mr.
Frans Timmermans (MP and spokesman in international issues) 990223
52
Compare the above description of the current procedures in PvdA, with the description in Koole, R. and
Leijenaar, M. (1988), aa, pp 197ff.
20
changes. Today, the ambition is to have a profile of the list that equals the different groups in
society. Before a career within the party was more crucial. However, these changes are highly
debated in the party.
The candidate selection in GL (Groen Links) is similar to the one in PvdA and CDA, although
there are some significant differences53
. The process begins with an adoption of a profile by the
party board. They also appoint a selection committee in the same way as the other two Dutch
parties with approximate 7-8 members. The local branches can nominate candidates and so can
policy specific groups within the party. However, the selection committee does most of the
work, according to Mrs. Cornelissen (one of the members of the selection committee for the EP
election). The nominated candidates have to fill in a form that inform the selection committee if
they are available for the assignment and give a short story of their experience, skill and political
views.
The selection committee starts its work by interviewing and scouting candidates. The committee
is responsible to fulfil the gender quota, which is regulated in the party statutes. They also need
to take into account the regional representation and certain knowledge that the candidates
should have. They also strive for including ethnic minorities on the list.
The proposal from the selection committee goes straight to the congress and not to the party
board as in the other two Dutch cases. This means that it is the selection committee that is
responsible to the congress and not the party board. According to Mrs. Cornelissen, the
committee works pretty much without any interference from the party board. The party board
can nominate candidates, but otherwise the committee keeps a high integrity in their work.
The congress makes the final decision and it is not unusual that they make changes in the
proposal from the committee. The reason is, according to Mrs. Cornelissen, that there is a
history of opposition within the party. It is not uncommon that groups within the parties
organise campaigns for candidates. Also this time, the congress made some changes in the list
proposed by the selection committee. The candidate in the third place on the proposal was
pushed down to fourth place. The committee proposed a newcomer, but the congress preferred
a more experienced candidate.
The congress makes the final decision and no other organisational body can veto the decision,
and there is no difference between how the party organises the selection of candidates to the
national parliament and to the EP.
The differences among the Dutch parties are small. In all parties, the national board adopts a
profile. To some extent this is done also in parties from other countries, but not as explicit as in
the Netherlands. Another similarity among the Dutch parties is that it is a congress that makes
the final decision. However, there are differences in how the proposal to the congress is
processed. All parties use a selection committee as a major component in the process. In CDA
and PvdA these selection committees are responsible to the party board, while in GL, the
committee gives their proposal directly to the congress. In CDA the proposal is forwarded to the
53
The material used for Groen Links is the party regulations and an interview with Mrs. Maria Cornelissen
(member of the selection committee for the EP-election) 990224 and Mr. Richard Wouters (Advisor on EUissues)
990224
21
local branches and they vote on the list. The result of this vote is the proposal for the congress.
In PvdA, the proposal from the party board goes directly to the congress.
The party board is therefore, a highly important actor in the selection of candidates in the Dutch
parties. Maybe, an exemption to this rule is the GL. The experience in PvdA is of interest in this
perspective. The party has centralised its procedure, as a way of monitoring the outcome of the
selection. This is also explicitly stated as a strategy to become a more popular party among the
electorate. The CDA still have an influence from the local branches, while in the GL the whole
process is in the hands of the selection committee.
The Netherlands is, as I have mentioned earlier, often categorised as having a centralised system
for candidate selection. This is also quite obvious in this study and if PvdA is some kind of a
forerunner, the system will be even more centralised in the future. However, much of this is due
to the design of the electoral system. A system with a single constituency naturally pushes the
selection process to the national level. Although, the parties have put forward several lists, based
on the provinces, this is not done in this EP election54
. Otherwise, there is no difference in the
formal procedures for the national elections and the EP-elections. However, several of the
interviewed representatives state that there are differences, although not formal. For example,
the conflicts surrounding the making of the lists are not as profound in the EP election as in the
national election. One reason is that the members are less interested in the EP election. One
other difference is that it is more difficult to make the lists for the EP election, because the
amount of electable candidates is smaller in the EP election than in the national election. It is
harder to fulfil different needs of the party when you have fewer representatives.
Comparison between parties according to the analytical procedure scheme
In table 3 below, I summarise the procedures in every party, according to the steps described
earlier. Now I use the analytical scheme as a way of analysing the similarities and differences
between the parties.
54
The process is not ended in some of the parties and still it is possible for the parties to add provincial
candidates before the election in June. However, none of the interviewed representatives believe that this will
be done in this election.
22
Table 3 EP Candidate selection procedure in social democratic parties, conservative
parties and Green parties in Ireland, the Netherlands and Sweden
Country and party Nomina
tion
procedu
re
Interme
diate
organis
ational
body
Influenc
e
Selectio
n
procedu
re
Binding Interme
diate
organis
ational
body
Influenc
e
Decisio
n
procedu
re
Ireland
LP 2 1 Yes 3 No 0 – 5
FF 2 1 Yes 3 No 0 – 5
GP 2 0 – 2 No 0 – 5
Netherlands
PvdA 2 2 Yes 5 No 0 – 3
CDA 2 2 Yes 5 No 1 Yes 3
GL 2 0 – 4 No 0 – 3
Sweden
SAP 2 1 Yes 5 No 0 – 3
MSP 2 1/3 Yes 2 No 3 Yes 5
MdG 2 2 No 2 No 2 Yes 3
Nomination procedure: 1=Free nominations, 2=Intra party nomination on constituency level or national level,
3=Subset of members on constituency level or national level, 4=Selection committee on constituency level or
national level, 5=National executive.
Intermediate organisational body: 0=No, 1=Regional/local level, 2=Selection committee, 3=National executive.
Selection procedure: 1=Inter party primaries, 2=Intra party primaries on constituency level or national level,
3=Subset of members on constituency level or national level, 4=Selection committee on constituency level or
national level, 5=National executive.
Intermediate organisational body: 0=No, 1=Regional/local level, 2=Selection committee, 3=National executive.
Decision procedure: 1=Inter party primaries, 2=Intra party primaries on constituency level or national level,
3=Subset of members on constituency level or national level, 4=Selection committee on constituency level or
national level, 5=National executive.
There are two types of intervening organisational body’s between the nomination procedure and
the selection procedure. The first one is when the nomination is processed through the different
levels in the organisation. The second type is when the party uses some kind of selection
committee. In LP and FF, the local branch is the intermediate body. It is also obvious that the
local branches have an influence in the process. Either they accept a nomination or they do not.
In SAP it is the regional level that collects the nominations. Also in this party the intermediate
body have an influence on the process. They can not prevent a nomination, but they decide
which candidate that shall represent the region in the continuing process. In MSP, the procedure
is different from the others. They use a preliminary primary among a subset of members on
national level and the national executive monitor the next step on the basis of this preliminary
primary. MSP is not a clear-cut case in my coding scheme. On the one hand the national
executive monitors this intermediate procedure, on the other hand representatives from the
regional level is involved in this step. All these parties; the two social democratic parties and the
two conservative parties are similar in this respect. They all use an intermediate procedure that
includes a subset of members on the regional or local level.
The second type of an intervening organisational body, the selection committee, is used in the
two major Dutch parties, PvdA and CDA. These selection committees collect nominations and
23
scout for candidates. They interview the nominees and compare them to the profile that the
party boards have adopted. They are both highly influential at this step in the process. The
Swedish green party also uses a selection committee. This committee collects the nominations
and organises the intra party primary. However, the selection committee is not able to intervene
in the process at this stage.
Two parties I have listed as not having this intermediate organisational body. In the Irish GP, the
nominations proceed from the members or local branches directly to an electoral convention on
constituency level. In the Dutch GL they have a selection committee, but this is responsible for
the whole process. In a way this is similar to the MdG. However, MdG uses an intra party
primary and I consider this to be the most important procedure for selection. This is not the case
in GL, therefor I do not consider the selection committee as an intermediate body, but as the
main locus for selection.
Non of the parties in the study uses inter party primaries as a tool to select candidates. The
voters have their saying on Election Day. However, the parties in the study use a variety of
procedures. Some of them include the members at this stage, but some does not. In SAP, the
party board makes the selection among the nominated and forwards a proposal for the decisionmaker
(Förtroenderådet). None of the Dutch parties includes the members or the local branches
at this stage. In PvdA and CDA it is the party board that makes the selection, after a proposal
from the selection committees. In GL, as I have already mentioned, it is the selection committee
that makes the selection.
In all other parties the members or the local branches are included at this stage in the procedure.
The Irish parties use selection conventions, where the representatives from the local branches
decide who will be the candidates. I have categorised the GP as having an intra party “primary”,
although the selection procedure involves a convention. The argument is that all members have
access to the convention and if they are not able to attend you have the possibility to give a
postal vote. This does not discriminate any member from being involved in the selection
procedure and therefore I categorise it as an intra party primary. MSP and MdG both use intra
party primaries.
In all the parties in the study someone else than the one that originally selects a candidate makes
the actual decision, after the selection is made. Some parties use a more or less important
intermediate organisational body between the selection and actual decision. Two types of
procedures are possible to identify. First, the one where the national executive or party board
has some kind of veto power. Secondly, where the party uses some kind of election conference.
Examples of the first type are the Irish parties the national party executives have a right to veto a
decision made by the selection conventions. Although several of the party representatives
interviewed in this study mean that this possibility shall not be used in the upcoming EP election,
the rule still exists and gives the national executive discretionary power. The Irish parties do not
have an intermediate organisational body at this stage. The national executive simply decides
upon the result of the selection convention. In the Swedish parties the picture is more diverse.
MSP is similar to the Irish parties. They involve the members in the selection stage, but give the
national executive veto power. Like in the Irish case, the MSP does not have an intermediate
organisational body. However, there is an important difference between the Irish parties and the
Swedish conservatives. The later has a result of a primary that they base their decision on. This
means that they probably can change the order of the list and add candidates more freely than in
the Irish parties. If the Irish parties, for example, add a candidate, the national executive would
24
present a conflict between candidates on constituency level, where the original decision was
taken and where there is an organisational body to defend the decision of the convention.
Hypothetically, this is not as a severe problem for MSP.
The second type of decision procedure, when the party arranges an electoral conference or
congress is used in SAP. The use of an electoral conference is a way for the party to legitimise
the selection of candidates, which up until this stage has been a comparably centralised process.
It is not necessary to have an intermediate organisational body with this procedure. The decision
of the party board goes straight to the electoral conference. MdG uses an extra congress, which
could be compered to a selection convention, but in this case their decision is binding. They also
have a selection committee as an intermediate organisational body. It is they who analysis the
result of the primary and propose the list to the electoral conference. The three Dutch parties
also use this second type of decision procedure. In the case of PvdA the proposal from the party
board goes directly to the congress. In CDA they use an intermediate level, when the local
branches vote on the list and the result of this vote is forwarded to the congress as the proposal.
In Groen Links the selection committee makes the proposal for the congress to vote on.
The analytical frame that I am using gives me the opportunity to study at what stage of the
process that the national leadership is influencing the process. At least in the formal aspects of
the procedures. In all parties included in this study, with the possible exception of the MdG and
GL, the procedures include some influence by the national executive body. The difference is
when this occurs. In some parties the national executive has some kind of veto over the decision
in the selection procedure. Let’s call this a post monitoring type of procedure. In other parties
the national executive operates the process by formulating the proposals for the more
decentralised regional body. Let’s call this a pre monitoring type of procedure. These are two
strategies for the leadership in the party organisations to have an influence of the selection of
candidates. In the pre monitoring type of strategy falls the Swedish SAP, the Dutch CDA and
PvdA. In the second post monitoring type falls the Irish parties FF, LP, GP and the Swedish
MSP. There are two cases that are difficult to put into these categories and that is the Dutch GL
and especially the Swedish MdG. As mentioned before it is hard to se that the party leadership
actually influences on the selection of candidates. In a way this have to do with their system with
a selection committee that are responsible for the whole process. This selection committee can
be more or less influenced by informal contacts with the party leadership, but I have nothing to
prove that this is the case. On the contrary, especially in the Swedish case, it seems like the
leadership puts a high trust in this selection committee and in the extra congress that makes the
final decision.
Conclusion
I put forward three main questions in the introduction of this paper. Are there differences in how
the parties from different party traditions and countries select there representatives? Secondly,
are there differences in how the parties organise there selection of candidates to the national
legislative body and to the EP? Thirdly, too what extent are the party leadership involved in the
process of selecting candidates to the EP?
The design of the electoral system has an effect on how parties organise their selection of
candidates. This has been showed in earlier studies and is verified in this study. The main
variation is between countries rather than within countries. The main component in the electoral
system is the number of constituencies. In Ireland the country is divided into four constituencies
in EP elections, while in ordinary national election the number is 41. This has a “pushing” effect
25
in the party organisations. They organise their selection of candidates in accordance with the
number of constituencies. This means that the process gets more centralised, or rather
regionalised, than in ordinary national elections. However, the main characteristics of the
procedures are maintained. This means that compared to other countries, Ireland, has a
decentralised candidate selection process as well as a decentralised electoral process. That the
number of constituencies has an impact on the party organisations is even more evident in the
case of Sweden. In ordinary elections the country is divided into 28 constituencies (even this is
large compared to the Irish case). In the EP election there is only one constituency. This makes
the procedure more centralised in the party organisations. Although the party organisations try
different means to legitimise the selection of candidates within the parties, the centralisation is
quite evident. In the Dutch case the number of constituencies are the same in the national
election and the EP election. Accordingly, the same procedures are used for the national and the
EP election. Although, several of the interviewed representatives mean that there are differences
in the interest from the members and the level of conflict between the national and the EP
election, the same formal rules are used. This means that there are differences in how parties
from different countries organise their selection of candidates. However, the main characteristics
are maintained although the number of constituencies. The exemption is Sweden, where the
change from 28 constituencies to one constituency has an effect on the parties’ procedure of
selecting candidates.
Sweden is the country where the parties use different means to select the candidates, i.e. Sweden
is the most heterogeneous example of the three included in the study. In the other two countries
there are minor differences among the parties. There can be different explanations to the
disparate picture in Sweden. The EP elections are a new phenomenon in Swedish politics. One
explanation is that the parties’ still searching for ways to handle the selection of candidates and
they try different means to do so. Another explanation could be that the Swedish political
context differs in one fundamental aspect the question of membership to the EU is still a highly
salient issue in Swedish politics, and the parties are divided within themselves on this issue. This
is the case especially in SAP, but it is also possible to see a rising conflict within MdG. In the
other countries there is a quite large consensus about the support of the EU and the countries
position on the European arena, within the parties as well as among the electorate. The
differences in procedures among the Swedish parties could be a result of the conflicts
surrounding this issue. The parties use different means to cope with the conflict and that is
reflected in the different procedures. It is hard to be too conclusive, with this limited study.
However, the effect of policy-conflicts within parties on the way they select their candidates, is
one aspect that should be further investigated.
With this small sample of party organisations it is not possible to conclude that the party heritage
have any strong explanation on how parties organise their selection of candidates. There are no
strong similarities between for example, the three social democratic parties. Afgain, it seems like
the country tradition and the electoral system has a more profound implication for the parties.
For example, there are more similar way to do this within the Dutch parties, than between CDA
and the MSP. However, one conclusion is important. The three green parties have slightly
different procedures than their national competitors. In the GP, the local branches do not have
the same importance as in FF and the LP and the selection procedure is more of a intra party
primary, than a subset of members in a convention. In the Dutch GL the selection committee is
more sovereign than in PvdA and CDA. This is also the case in the Swedish MdG, compared to
the other Swedish parties. Why is this? In the discussion about the Irish GP, I suggested one
possible explanation. These parties are smaller than the other ones. Perhaps they do not need
26
procedures of check and balance as in the other larger parties. For example, the need for the
national executive to intervene in the process or the local branches to monitor the selection. It
may be that this is not as complicated story in these parties as in the larger ones. On the basis of
this study, it is of course, difficult to say. However, this study proposes that it is meaningful to
analyse these parties as a separate category.
Now then, is the candidate selection enterprise a highly centralised guided procedure in the
modern party? There seems two be two ways for the national executive to intervene in the
process of candidate selection in the parties included in this study. I call these pre or post
monitoring strategies. It is of course discussible which one of these strategies that give the
leadership most space for manoeuvring, but I am prepared to argue that the pre monitoring
strategy gives the leadership the most powerful tool to influence the selection of candidates. It is
a question of agenda setting (or rather a ticket setting). If the leadership of the organisation is
the one that proposes a certain disposition of the list, it is quit difficult for the more
representative body of the organisation to alter this proposal. Although, there is a hypothetical
chance, the votes outbalance each other and it is much easier to defend the whole composition
of a list, than make changes of the position of a certain candidate. However, this conclusion
deserves more empirical research. This research should also include the extent to which the
selected and elected candidates remain linked to their party organisations and their
constituencies when they leave for Brussels and Strasbourg.
27
Written references
Allen, K. (1997), Fianna Fail and Irish Labour – 1926 to the present, Pluto Press, London
Bryde, T. (1998), Party Groups in the European Parliament and the Changing Recruitment
Patterns of MEP’s, in Bell, D.S. and Lord. C. (1998) Transnational Parties in the European
Union, Ashgate, Hampshire
Budge, I. (1996), Great Britain and Ireland – Variations on Dominant Party Government, in
Colomer, J.M. (1996) Political Institutions in Europe, Routledge, London
Busteed, M.A. (1990), Voting Behaviour in the Republic of Ireland, Clarendon Press, Oxford
Bäck, M. and Möller, T. (1997), Partier och Organisationer, Norstedts Juridik AB, Stockholm
CDA regulations for the EP candidate selection, ”Reglement voor de voorbereiding van
kandidaatstelling voor de verkiezing der leden van het Europese Parlament”,
Chubb, B. (1992), The Government and Politics of Ireland, Longman Group Limited, London
Cox, G.W. (1997), Making Votes Count, Cambridge University Press, Cambridge.
Dagens Nyheter, 990110, Kritiker bör inte ställa upp i EU-valet.
Dagens Nyheter, 981215, EU-Fronten i mp rämnar.
Dagens Nyheter, 990109, Schori lanseras som kandidat till EU-valet.
Dagens Nyheter, 991222, Mp-krav på”ut ur EU” står fast.
Dagens Nyheter, 991229, EU-positiva petas från mp:s EU-vallista.
Farrell, B. (1985), Ireland: From Friends and Neighbours to Clients and Partisans: Some
Dimensions of Parliamentary Representation under PR-STV, in Bogdanor, V. (1985),
Representatives of the People?, Aldershot, Glower
Farrell, D.M. (1994), Ireland: Centralization, Professionalization and Competetive Pressures, in
Katz, R.S. and Mair, P (ed), How Parties Organize, Sage Publications Ltd, London
Gallagher M, Laver, M and Mair P (1995), Representative Government in Modern Europe,
McGraw-Hill Inc, New York
Gallagher, M. and Marsh, M. (ed.) (1988), Candidate Selection in Comparative Perspective –
The Secret Garden of Politics, Sage Publications Ltd, London
Generaldirektoratet för Forskning, Lagarna För Val Till Europaparlamentet, Politiska Serien
No.13, Europaparlamentet 1997
Gladdish, K., The Netherlands, in Bogdanor, V. (1985), Representatives of the People?,
Aldershol, Glover
Green Party is the statutes from 1997
28
Groen Links party regulations
Irish Political Studies, vol. 8, 1993, pp 202
Irish Political Studies, vol. 9, 1994, pp 231
Johansson, K-M. (1996), The Nature of Political Parties in the European Union, in Kourvetaris,
G. A. and Moschonas, A. The Impact of European Integration, Preager Publishers, Westpoint
Katz, R.S. (1980), A Theory of Parties and Electoral Systems, John Hopkins University Press,
Baltimore
Katz, R.S. and Mair, P. (1994), How Parties Organize, Sage Publication Ltd, London
Katz, R.S. and Mair, P. (1995), Changing Models of Party Organization and Party Democracy,
Party Politics, Vol 1, pp 5-28
Katz, R.S.and Mair, P. (1992), Party Organizations: a datahandbook on party organizations in
western democracies, 1960–1990, Sage, London
Koole, R. (1994), Ireland: Centralization, Professionalization and Competetive Pressures, in
Katz, R.S. and Mair, P (ed), How Parties Organize, Sage Publications Ltd, London
Koole, R. and Leijenaar, M. (1988), The Netherlands: the predominance of regionalism, in
Gallagher, M. and Marsh, M. (ed.) (1988), Candidate Selection in Comparative Perspective –
The Secret Garden of Politics, Sage Publications Ltd, London, pp 201
Labour Party ”Standing orders for European Parliament selection convention”
LeDuc, L., Niemi R.G. and Norris, P. (1996) Comparing Democracies, Sage Publications Ltd,
London.
Mair, P., Party Organizations, Vote Management, and Candidate Selection: Toward the
Nationalization of Electoral Strategy in Ireland, in Penniman, H.R. and Farrell, B. (ed) (1987),
Ireland at the Polls 1981, 1982, and 1987, Duke University Press, Durham N.C.
Marsh, M. (1995), The 1994 European Parliament Election in the Republic of Ireland, Irish
Political Studies, Vol. 10, pp 209-215
Marsh, M. The European Parliament Election in the Republic of Ireland, Irish Political Studies,
Vol. 10, 1995, 209-215.
Müller, W.C. (1994) The Development of Austrian Party Organizations in the Post-war Period,
in Katz, R.S. and Mair, P. (1994), How Parties Organize, Sage Publication Ltd, London
Norris, P. (1996) Legislative Recruitment, in LeDuc, L., Niemi, R. And Norris, P. (1996)
Comparing Democracies, Sage Publications Ltd, London
Panebianco, A. (1988), Political Parties: Organization and Power, Cambridge University Press,
Cambridge
Pitkin, H.F. (1972), The Concept of Representation, University of California Press, Berkeley
29
PvdA party statutes from 1998
Rae, D.W., (1975), The Political Consequences of Electoral Laws, Yale University Press Ltd,
London
Skare, A (1996) Kandidatutvelging – mer enn riktig kjønn fra rett sted, Tidskrift for
samfundsforskning, Vol 37, no 3, pp 331
Swedish Conservative Party is the party statutes
Swedish Green Party statutes from 1992 party congress (revised at 1993 and 1994 party
congress)
Swedish Social Democratic Party party statutes from 1997 party congress
Svenska Dagbladet, 990131, Schori hotar Theorins EU-plats
The Irish Times, 980624, Invitation to Cassells causes Labour row.
The Irish Times, 990208, Gallagher selected for Euro election.
Interviews
Christen Democratisch Appel: Mr. J.M. Wiggers (head of the International Secretariat) 990222,
and
Christen Democratisch Appel: Mr. MJM Verhajen (MP, spokesman on international issues and a
former MEP) 990223
Fianna Fáil: Dr. Rory O’Hanlon (Leader of the Parliamentary Party and Deputy Chairman of the
Dáil Éireann) 990211
Fianna Fáil: Mr. Martin Mackin (general secretary) 990209
Green Party: Mr. Phil Kearney (co-ordinator) and Mr. John Goodwillie (rules and procedure
committee) 990210
Groen Link: Mr. Richard Wouters (Advisor on EU-issues) 990224
Groen Links: Mrs. Maria Cornelissen (member of the selection committee for the EP-election)
990224
Labour Party: Mr. Ray Kavanagh (General secretary) 990208
Miljöpartiet de Gröna: Kjell Dahlström (General Secretary) 990120
Miljöpartiet de Gröna: Marianne Samuelsson (Parliamentary party group leader) 990121
Moderata Samlingspartiet: Cleas Löfgren (party board) 981102
30
Moderata Samlingspartiet: Gunnar Hökmark (General Secretary) 990203
Partij van de Arbeid: Mr. Arjen Berkvens (Assistant international secretary and a member of the
PES-delegation) 990224
Partij van de Arbeid: Mr. Frans Timmermans (MP and spokesman in international issues)
990223
Socialdemokratiska Arbetarpartiet: Mr. Stefan Svensson (national organisational chief) 981118
and 990119
Socialdemokratiska Arbetarpartiet: Sven Hulterström (Parliamentary party group leader)
990122
31
Nomination
procedure
Selection
procedure
Decision
procedure
National
executive
National
executive
National
executive
National
executive
Electoral
committee on
constituency
level or national
level
Electoral
committee on
constituency
level or national
level
Electoral
committee on
constituency
level or national
level
Party
members
Subset of
members on
constituency
level or national
level
Intermediate
organisational
body
Subset of
members on
constituency
level or national
level
Intermediate
organisational
body
Subset of
members on
constituency
level or national
level
Intra party
nomination on
constituency
level or national
level
Intra party
primaries on
constituency
level or national
level
Intra party
primaries on
constituency
level or national
level
Voters Free nomination Primaries Primaries
Procedure scheme of the selection of candidates
Appendix 1