i
j
i.'!
i,,I
I j,
>1
i
J
i
Ir;
1 PUZZLES IN POLITICAL
RECRUITMENT
This book aims to provide the first full account of legislative recruitment in
Britain for twenty-five years. The central concern is why some politicians
succeed in moving through the 'eye of the needle' into the highest offices of
state. In democracies, many participate as grassroots party members) community
activists, and campaign donors. Some become local or regional
elected officers, party leaders, or lobbyists. From this pool of eligibles, some
run for parliament, a few are elected, and even fewer rise into government
office. Recruitment operates for offices at all levels. Legislatiue recruitment
refers specifically to the critical step as individuals move from lower levels
into parliamentary careers. In practice, given the accidents of political life]
many careers are far from linear. During their lifetime] politicians may
transfer laterally] skip a step or two along the way] or move up and down
offices, like a game of snakes and ladders. This study of legislative recruitment
explores how and why people become politicians] and the consequences
for parties] legislatures and representative government.
Many different perspectives within political science provide insights into
this cornman concern.' Among the most traditional approaches] a rich
biographical and historical literature documents the careers of political
leaders based on memoirs] letters and public records. Early sociological
- theorists such as Mosca] Pareto and Michels] and neo-marxists, were concerned
about the outcome of the recruitment process] the way legislative
elites restricted access] and their privileged class origins," More recently]
increased concern bas been expressed about the barriers to entry facing
-women and ethnic minority candidates.P Institutionalists interested in party
organisations have studied the decision making process over candidate
selection] for the insights this provides into the distribution and
. centralisation ofpower within parties. Following Lasswell] political psychologists
concerned with political motivation sought to identify a distinctive
personality among lawmakers] which drew them into public life," Psepholo-,
gists have concentrated on the electoral consequences of candidacies]
notably the 'personal vote' incumbents may atrract.? Rational choice theo"rists
have sought to model the decision to run] based on the perceived costs
Studies of party organisations: who selects and now?
Parties serve vital functions linking citizens with government: they structure
electoral choice, provide a legislative agenda for government) and recruit
legislative candidates. Candidate selection may seem at first sight like a
routine and obscure function of political parties, conducted behind closed
doors in small meetings long before the public drama and excitement of the
election campaign. In marginal seats) who gets into parliament is determined
by voters. But in safe seats with a predictable outcome the sclectorate
have defacto power to choose the MP. And in Britain, about three quarters
of all seats are 'safe, with majorities greater than 10 per cent.? In choosing
candidates the selectorate therefore determines the overall composition of
parliament, and ultimately the pool of those eligible for government. In
federal systems such as in Canada or the United States, there are multiple
routes into government. But in Britain there is a single ladder into the
highest offices of state; the first hurdle is adoption as a prospective parliamentary
candidate in a local constituency.!"
and benefits of different levels of office.6 Legislative specialists have studied
how the background andcareers of politicians relate to their activities and
roles, and the consequences of candidate selection for parliaments."
Recruitment studies stand at the intersection of research on mass political
participation, elections and voting behaviour, political elites, legislatures,
parry organisations, and interest groups, as well as, more recently, gender
and racial politics. This intellectual diversity exerts centrifugal pressures
which tend to fragment recruitment studies, as each perspective emphasises
different theoretical frameworks, key questions and methodological
approaches. BBut the potential ability to draw on many subfields can also be a
source of considerable intellectual strength.
To understand recruitment, this study seeks to reintegrate the literature
from two primary subfields in political science. Studies of political elites
have been concerned with the social composition of parliament. Studies of
party organisations have focused on how the process operates and what the
selection process tells us about the distribution of power within parties. This
book seeks to build on this literature, developing a more comprehensive
theoretical model and analysing new evidence - the British Candidate study
(BCS). The aim is to link our understanding of the process of Candidate
recruitment with the outcome for the social composition of parliamentary
elites. This study provides a fresh exploration of three major questions:
(i) Who selects, and how?
(ii) Who gets selected, and why?
(iii) Does the social bias of the outcome matter?
3Puzzlesin political recruitment
In Ranney's words, factional struggles to control the nominating procedure
"are contests for 'nothing less than control of the core of what the party
.. stands for and docs'." Placing candidates in safe seats, possibly for a
{'lifetime political career, has more significant consequences than getting
':':'.:·conference resolutions adopted, or supporters nominated to internal party
cbodies. In the Conservative party, disputes over nominations have usually,
/'but not always, been resolved behind closed doors. In the Labour parry,
.~~Iactions have struggled more publicly to control the selection process. In
,,,1993 this was vividly illustrated by the heated Labour party debate about
<,~ihe appropriate powers of trade unions versus grassroots parry members]
,vnth the conference argument over (one member - one vote' which almost
·,:brought down the leadership.
:-_ 'e-The locus of control over candidate selection varies substantially cross'~":-:'ationally.
In most countries the recruitment process is governed primarily
-y. internal party rules, rather than by law.!? A comparative approach
indicates that decision making in the recruitment process varies along two
./'dimensions. First) there is the question of the dispersion of power. Is the
The main approach to studying recruitment in Britain has focused on
identifying who controls selection decisions within parties; whether
national leaders, local officers, or grassroots party members, and how this
power has evolved over time. Studies have documented the basic steps in
the selection process, and what this tells us about the distribution of power
within parry organisations. The recruitment process has commonly been
evaluated according to whether the process is 'democratic' in the sense of
involving local activists and grassroots members; 'fair' in treating all applicants
equally; 'efficient' as a decision making process; and 'effective' in
producing 'good' candidates. The appropriate weight given to these criteria,
and whether the system meets these objectives, have been subject to heated
debate.
The question of internal party democracy, particularly the appropriate
role for national and local organisations, has been one of the most controversial
issues. Ever since publication of Ostrogorski's classic work at the turn of
the century,I
1
studies have been interested in who has, and who should have,
.. control over selection, comparing the role of the national party leadership,
:i·'.,.,-local constituency officers) party factions) and grassroots party members.
Struggles to control the process have always been one of the prime areas of
intra-party conflict, as Schattschneider notes, because gatekeepers who
.select ultimately control the composition of the parry leadership:
The nominating[i.e. candidateselecting] process .. ,has become the crucialprocess
of the parry. The nature of the nominatingprocedure determines the nature of the
party;he who canmakenominations is the ownerof the pany.ts
Political recruitment2
lative recruitment in
vhy some politicians
the highest offices of
tarry members, comne
local or regional
ool of eligibles, some
rise into government
~gislative recruitment
rve from lower levels
:lents of political life,
nne, politicians may
r move up and down
of legislative recruiticians,
and the con-
government.
: provide insights into
II approaches, a rich
e careers of political
:ls. Early sociological
o-marxists, were canis,
the way legislative
igins.2 More recently,
.rriers to entry facing
ists interested in party
recess over candidate
the distribution and
swell, political psycho)
identify a distinctive
public life.' Psepholoences
of candidacies,
j Rational choice theoon
the perceived costs
process centralised with the main decisions taken by the national party
leadership, is it left to regional party officers, or is it dispersed with
grassroots local parry members exerting most influence? Secondly, there is
the question of the formalisation of decision making. Is the process
informal, a matter of tacit norms with few binding rules and constitutional
regulations, or it is formalised so that the procedures at each step are
standardised, rule-governed and explicit. These distinctions suggest six
main types of selection process (see figure 1.1).
In informal-centralised systems (such as the French Union pour la Democratie
Francaise - UDF) there may be democratic constitutional
mechanisms, but in practice the process is characterised by leadership
patronage. Rules serve a largely symbolic function. Without any established
tradition of internal party democracy, and with loose organisations, party
members play little role in the process. In informal-regional systems (such as
the Italian Christian Democrats) faction leaders bargain with each other to
place their favoured candidates in good positions.IS
In informal-localised systems (such as in the Canadian Progressive Conservatives),
local ridings decide on the general procedures used for selection,
as well as the choice of individual candidate. Without established
guidelines, practices vary widely; some constituencies may, nominate at
large-scale meetings open to all tmernbers", while patronage by a few local
leaders may be significant in others. Reflecting weak organisations, this
system may be open to manipulation by small groups.
Alternatively, informal-centralised and formal-regional systems (such as
in the Liberal party in the Netherlands, the old Italian Communist Party
(PCI), or the old Japanese Liberal Democrats), party executives or factional
leaders at national and regional level have the constitutional authority to
decide which candidates are placed on the party ticket. Lastly the most
common pattern in European parties is one of formal-localised recruitment.
Here constitutional rules and national guidelines are established to standardise
the process throughout the party. The fairness of the system,
ensuring all applicants are treated alike, rests on the implementation of
Figure 1.1 Decisionmakingagencies
5Puzzlesin political recruitment
clear, transparent and equitable rules. Within this framework the selection
of individual candidates takes place largely by local agencies at constituency
level.
":(.: ,Based on this classification, it becomes apparent that in the long term the
",:,"," main change in recruitment within British parties has been in process rather
'than power. There has been a gradual evolution from an 'informal-localised'
system based on patronage in the nineteenth century towards a more
'formal-localised' system today based Onmore meritocratic standards. This
o. Change has gone further in some parties. than others. At the turn of the
;>:century Ostrogorski provided one of the earliest accounts of the trans~;:~
.,"fo~atianof this system.Is In mid-Victorian Britain, local patronage predo'"L":~ated;
a few local notables would throw their weight behind candidates
::.with sufficient independent resources and social connections for an effective
,::c:ampaign. The 1832 Reform Act led to the development of more formal
(~egistration Societies. At local level the first Conservative associations
,:r:~eveloped during the mid-1830s to bring in regular subscriptions, organise
~,~electoral registration, and rally electoral supporters.'? At national level the
Rg'reatpolitical clubs - the Carlton and the Reform - provided a rudimentary
Tp.~ organisation, functioning as a social base bringing together poli"::~cta.~s,
party agents, local associations and influential supporters from the
§lt~Vll1Ces. Formal party labels meant little, after the split over the Corn
:~ws"when there were shifting parliamentary factions based around poli"~calleaders.18Ostrogorski
was concerned with the development of modern
parties from small, informal factions into structured mass-branch organi'"
tIons following expansion of the franchise, the introduction of the secret
iallor, and reform of corrupt practices.'?
{The 1867 Reform Act provided the major catalyst for the organisation of
ass parties. The Liberal party was transformed by Joseph Chamberlain's
;>~ation of the 'Birmingham Caucus' in 1867, and the subsequent develop,ent
of the National Liberal Federation in 1877. The Conservatives were
FIarly transfigured by the creation ofthe National Union in 1867 to bring
_,getherthe constituency associations, the creation of Central Office in 1870
~e;;:".coordinate the professional services, and Lord Randolph Churchill's
,,~"?rganisationof the National Union in 1886. From its earliest beginning,
~-,~:,organisationof the Conservative party outside of parliament Was con"yed
'as a servant of the party within parliament. Local branches were
'i~blished as election machines, to mobilise the newly enfranchised voters
o-became "too numerous to reach by traditional means.20 Reflecting their
;~nger orgarnsation,constiruency associations were given two new funcn~:
to enable supporters to influence the party programme, and to provide
.ore. popular body for selecting candidates.s! Ostrogorski's central
ety-was the effect of 'caucus control'; if MPs became accountable to
LocalRegional
Informal :
process .
----
Formal
process
'------
Central
Political recruitment4
j
iit
'j
i
'I
1~"
rank-and-file party members) he feared this' would undermine the independence
of members of parliament.
Following in his footsteps) McKenzie's authoritative study of British
parties in the mid-fifties established that Ostrogorski's fears of caucus
control were groundless.F McKenzie found that Conservative and Labour
constimency associations had considerable autonomy over whom they
adopted within certain agreed rules. Nevertheless) once elected legislators
were rarely accountable to local members. So long as they remained 'en
rapport' with their constituency parry) McKenzie concluded that British
MPs could act as Burkean trustees, able to exercise independent judgement
over issues. Due to the deference of parry members, MPs rarely functioned
as delegates mandated by local activists.P There were few cases of constituency
de-selection of elected members. Nevertheless) the growth of
organised parties did undermine the independence ofMPs. Party discipline
was applied directly through party whips and national officers, who could
threaten the ultimate punishment; official withdrawal of the party label.
24
In short, McKenzie concluded that constituency powers over the selection
process did not lead, as some had feared) to direct local control over MPs)
although national party control increased.
This established the textbook wisdom for many years. Following in this
tradition, the major books on the recruitment process in Britain) pUbIis~ed
in the 1960s by Austin Ranney and Michael Rush, were concerned WIth
documenting the main steps in the selection process.25 The focus was on
identifying the influence of key actors and analysing sources of potential
conflict between central party headquarters and local activists. The studies
outlined the rules, examined the social characteristics of candidates on the
basis of aggregate data, and compared case studies in some constituencies.
The selection of candidates) the authors confirmed) remained the prerogative
of local parties, with the main decisions in the hands of constituency
officers. Indeed, this was one of the few areas where local parties remained
largely autonomous. The outcome of the process - why some candidates
were selected over others - was treated as an issue with few conclusive
answers.
At the same time) Peter Paterson produced a strong case for reform,
arguing that undue power rested with secretive and unrepresentative party
cliques." Influenced by the movement against caucuses in the United
States, Paterson felt that small selection committees in Britain needed to be
replaced by democratic party primaries, open to all members. This proposal
was supported in the mid-1970s by the Hansard Society." Subsequent
work has focused on the causes and consequences of bitter internal splits
over selection battles within the Labour party)28 and left wing moves to
introduce mandatory reselcction, in an attempt to make the parliamentary
7Puzzles in political recruitment
Changes in the Selection Process
.i~en this literature there are several reasons for a fresh look at the
"j1iitment process. First, there is a need to establish how the process has
perated in recent years, taking account of changes over time. Observers of
,_:!:ection meetings today, reading accounts of the 1950s and 1960s) would
"~6gnise much that is strikingly familiar.!' In time-honoured fashion
.didates continue to apply for particular constituencies, undergo a
cess of interview and short-listing by local party bodies, until one
ornes.the official party standard-bearer. Nevertheless, during the last
'.?es many aspects of the Labour and Conservative selection process
-~_·changed significantly. Reforms have usually_been initiate~Lduring
flds'in opposition, when parties hav.!....§ougltt.!Q..I£gain e~f!2E~1 popular..~Y:u-lii?ro~g
the quality oCtl1e!!:-~andiE.~tes..T:f.1e,_~~I~,c:tion,pro-~e5S1lliS
r~~.§·,~~~_~.oraancewIili::~:"..~~~~~E!._~mq_s ..a~itraditio~-aT.p~iGtL~~lLm
parry accountable to Labour activists.P? The debate about the appropriate
influence of trade union affiliates over the choice of Labour candidates
follows this tradition.P? and proposals for greater internal democracy with
'one-rnember-one-vote' echo back to Paterson.
Previous studies established the characteristic 'formal-localised' nature of
the recruitment process as it operated in British parties during the 1950s
and :19605. During this period constituency associations - mainly core
and affiliated factions - made the major decisions about the choice
of individual candidates. At the same time the national leadership determined
the general rules, supervised the process, and exercised formal veto
Dowers. to ensure that the process was fair and efficient.
:.~,b,,:It was commonly' assumed that a formal-localised system was functional
;:;~..f9r,~_British party organisations. \Vithout some central management the
~::-P!O:c;ess might become factionalised and divisive) since in moribund con:::,:~ptuency
associations small groups might 'capture' the party label for their
f:preferred candidate. Standard procedures for selection and appeal help
,:H-~sure that the rules are seen as uniform and legitimate by all participants.
:;:;':¥lBritish parties) except the Greens, have national guidelines, and formal
ciw¢tting of all proposed candidates by national officers. On the other hand it
b):~~usually assumed that too much control by the national party leadership
tpight cause resentment at the grassroots level. The constituency association
,;-.~s·:,:;to work closely with their candidate on a day-to-day basis for an
~ffective grassroots campaign. Local members are most in touch with the
P:.~.edsof their area. Therefore, many believe that local associations should
;~~ercise most power over the choice of individual applicants, working
~~tb.in nationally standardised selection rules.
Political recruitment6
the organisation of
ieph Chamberlain's
ubsequent developConservatives
were
.ion in 1867 to bring
enrral Office in 1870
ndolph Churchill's
; earliest beginning)
tarliament was conocal
branches were
enfranchised voters
5. 20 Reflecting their
iven two new funcrune}
and to provide
strogorski's central
arne accountable to
5
[1 the long tenn the
:0 in process rather
informal-localised'
y towards a more
tic standards. This
At the turn of the
unts of the trans':"
IIpatronage predobehind
candidates
ons for an effective
mt of more formal
vative associations
scriptions, organise
.t national level the
ided a rudimentary
ring together poliupporters
from the
rplit over the Corn
based around polilopment
of modern
lass-branch organiuction
of the secret
twork the selection
.ies at constituency
each party. In the major parties the main impact of these changes has been
two-fold: to increase the formality of the process; and to shift power
slightly away from the core constituency activists, simultaneously upwards
towards the central leadership and downwards towards grassroots
members.
As described in chapter 3, the Conservatives revised the 'model' rules
guiding procedures following the Chelrner report in 1972, slightly strengthening
the role ofparry members at the expense of the constituency executive
committee. In 1980, Conservative Central Office introduced manageriaIist
selection boards to scrutinise the pool of eligibles on the Approved List
before they could apply to particular constituencies. These boards were
designed to produce better quality candidates and a meritocratic, open and
fair system. This legitimised and thereby strengthened control over the pool
of eligibles by Central Office. At the same time the Conservatives tried to
make sure grassroots members in general meetings had a genuine choice of
finalists.
Labour changed its rules during the 1980s, as part of the general process
of party modernisationJ
described in detail in chapter 4. Driven by conflicting
internal pressures, Labour implemented mandatory reselection for
incumbentMPs, formalised the selection procedure, shifted power downwards
from the constituency General Management Conunittee to an
electoral college of all members, and allowed greater NEe intervention in
the choice of by-election candidates. To encourage more women candidates,
Labour altered the shortlisting rules, expanded training programmes, and
has recommended the use of all women shortlists in half the seats where
Labour MPs retire and half the Labour target rnarginals, although it
remains to be seen whether there will be legal challenges, and if and how this
will be implemented. The role of trade union sponsorship, and the power of
union affiliates over selection, became subject to increased criticism in the
1980s. In October 1993 Labour decided to move towards a one-memberone-vote
selection system, with trade unionists participating as individual
members.
During the last decade, innovations have also been introduced in the
minor parties. The SDP/Liberal Alliance, subsequently the Liberal
Democrats, developed training programmes for candidates, introduced
postal ballots for members, and initiated shortlisting quotas for wom~n.
The decentralised Greens adopted more informal procedures; local parnes
largely determine their own procedures, and all applicants stand for
constituency hustings, unlike other parties there being no prior process
of shortlisting. The Scottish National parry uses a fairly rigorous series ~f
exercises to establish whether applicants can be placed on their
approved list.
9Puzzles in political recruitment
The sociology of political elites: who gets selected]
and why?
study of parry organisations focuses on how the process operates and
'.- -haspower over recruitment. This perspective can be understood as one
~of the equation. It is supplemented by the extensive literature on
.,rticalelites, concentrating on the outcome of the process. The traditional
:}?logical study of political elites sought to explain how those in power
,:,. arced and consolidated their position.P" Robert Michels provided the
':est theoretical account of how parry leaders exercised control over
aroots members through the 'iron law of oligarchy'J even in parties like
;,;;~erman Social Democrats which officially subscribed to notions of
"party democracy.
est of the empirical work on political elites in Britain has been con·~d';with,
documenting trends rather than with explaining the com'~,on
,of parliament in terms of the process of recruitment. That is,
..:..:es have focused on who got into positions of power rather than how they
The attitudes of parry selectors
The second reason for a fresh study lies in the need to move beyond the
formal process to analyse the attitudes, values, and priorities of parry
.selectors. The continuing puzzle is to understand why some are chosen over
others. Bochel and Denver produced a path-breaking survey of the attitudes
'C'_-' .of party selectors in the Labour party in Scotland and the north.P This
:L->survey was innovative but limited in scope, and has not been replicated.
-Recent surveys of party members allow the .first systematic analysis of the
>:_political behaviour of grassroots activists.P But, somewhat surprisingly,
;~:.these studies did not gather information about the experience of party
.members in the candidate selection process, or members' attitudes towards
::'.;;:::their elected representatives. These surveys have been limited.to comparing
).members and voters. Without a broader theoretical model of representative
~>;~~,~ocracy they have not envisaged activists as a middle stratum linking
',;:'~rectors and MPs. The institutional fbcus of organisational studies means
-,::;:vie-know more about the main steps in the process than the experience and
..-'~ttitudes of the key actors. What are selectors looking for in candidates,
~~en they make their decisions? Do participants feel that selection pro.::~ures
are fair, democratic and efficient? Are parry members and candi:q,ates
satisfied with the process? What do members feel about the relative
:~uenceof national and local party agencies? To understand the experience
~dperceptions of the main actors we need to go beyond the formal steps in
e process.
Political recruitment8