Scandinavian Political Studies, Vol. 26 ­ No. 4, 2003
ISSN 0080­6757
 Nordic Political Science Association
307
Mid-Level Officials as Policy Makers:
Anti-Corporatist Policy Change in the
Swedish Employers' Confederation,
1982­1985
Joakim Johansson*
The aim of the article is to provide an account of the process of policy change in the Swedish
Employers' Confederation (SAF), 1982­1985, implying a shift from a pro-corporatist to an anticorporatist
view on interest representation within decision-making bodies of public authorities
in the sphere of labour market and working life issues. In this respect, the study throws new light
on the fall of the `Swedish modeľ of industrial relations, by stressing the central policy-making
role of a few individuals occupying their positions in the mid level, rather than in the leadership
level, of the huge SAF hierarchy.
Given the formal structure of SAF and its statutes, SAF seems to be the least likely organisation
to show signs of policy making in the mid level. Therefore, the case study also contributes
to the general discourse about policy making in organisations, foremost by challenging mainstream
rational choice theory assumptions of the role of the formal leadership in processes of
policy change.
Being based on studies in the archives of SAF, the article reveals the mechanisms explaining
why a few mid-level officials were successful in anchoring a minority standpoint into the basically
`model-friendly' leadership of SAF. The argument put forward is that the key to an understanding
of this case of minority influence is to consider the mid-level officials' strategic use of
different kinds of information-based persuasion and propaganda techniques. In fact, the policymaking
mid-level officials belonged to a specific activist subgroup within SAF with its main base
in SAF's department of information.
Introduction
In 1985 the Swedish Employers' Confederation's (SAF's) board of directors
decided to recommend the abolition of interest organisations' right to be formally
represented in decision-making bodies of government agencies. The
decision was a serious blow against the `Swedish modeľ of institutionalised
three-party cooperation among labour, capital and the state on labour market
and working life issues. In 1991 SAF decided to withdraw unilaterally
* Joakim Johansson, Department of Government, University of Uppsala, Box 514, 751 20 Uppsala,
Sweden.
308
from the National Labour Market Board (AMS), the National Board of
Occupational Safety and Health (ASS) and a large number of other government
agencies, and in 1992 the bourgeois majority in the Swedish parliament
decided to change the composition of these bodies, referring to the withdrawal
of SAF. The latter decision meant that the central trade union organisations,
the Swedish Trade Union Confederation (LO) and the Swedish
Confederation of Professional Employees (TCO), lost the right to be represented
in the agencies and on the boards from which SAF had withdrawn.
As a critical juncture the decision taken by SAF in 1985 seems a very
important one to study. The aim of this article is to provide an account of the
policy change in SAF, and by doing so to throw new light on the fall of the
Swedish model by stressing the central role played by a few individuals occupying
positions in the mid level, rather than the leadership level, of the huge
SAF hierarchy. Thus, I challenge the vast amount of academic literature
about SAF and the fall of the Swedish model of industrial relations, which
explicitly or implicitly tells us the story of an essentially leadership-driven
process of policy change in SAF (De Geer 1989; 1992; Blyth 2001),1
sometimes
by combining this view with an account of the central role played by
some of the SAF federations (Swenson & Pontusson 2000; Pontusson 2002).
There are also general lessons about policy-making processes in organisations
to be learnt from this case study of SAF. Given the formal, indeed very
hierarchical, structure of the organisation and its formal statutes, SAF seems
to be the least likely organisation to show signs of policy making at the mid
level.2
Evidence of policy making at the mid level in such an organisation
could therefore contribute to our general body of knowledge about policy
making in organisations.
Theoretically, the argument is put forward within the framework of the
classical Cyert and March (1963) perspective on organisations. Cyert and
March describe organisations as consisting of loosely coupled functional
units, giving rise to subgroups, often with interests quite different from those
of the formal leadership. Being consistent with an assumption of utilitycalculating
individuals, this perspective on organisations is basically different
from those of some subsequent theories of organisational decision making
such as the `garbage can' or the `logic of appropriateness'. However, in contrast
to most rational choice models of organisational decision making this
perspective doesn't take the formal leadership's ability to steer organisations
for granted. Rather, it contains reasons to put such an assumption seriously
into question, or at least to make the role of the leadership an empirical issue.
Next, I will briefly introduce the cultural context in which the policy
change took place. After that, I turn to the case study, structuring my analysis
around the concept of four different phases in the policy-making process:
a phase of initiation (i.e. when corporatist arrangements in government
bodies within the labour market and the working life sector were for the first
309
time considered problematic by people in SAF), a phase of investigation (i.e.
when the issue was investigated within the organisation), a phase of decision
making (i.e. when the organisational leadership had to choose a standpoint
on the issue on behalf of the organisation) and finally a phase of implementation
(i.e. when the anti-corporatist standpoint was communicated to actors
in the political system external to SAF).3
In the last part of the article, I sum
up my results and highlight my conclusions.
Political Culture in SAF
During the 1980s SAF was a huge organisation employing some 500 people,
and also a quite heterogeneous organisation encompassing different cultures
with very different ideas about the appropriate role for SAF to play in the
political system. Three distinct cultures can be identified: the negotiators, the
deliberators/policy-experts and the activists.4
The former two cultures were the two main pillars of a dominating `modelfriendly'
SAF culture, with strong ties into the leadership of SAF. Emphasising
the virtue of mutual understanding among opposite parties (SAF
and the central trade union organisations), the negotiators had a profound
pro-corporatist view on politics, whereas the deliberators'/policy-experts'
approach was more pragmatic, stressing the expert role of organisations and
their task of broadening the basis of political decisions in a deliberative manner.
However, taking into account that much of the experts' daily work consisted
of representing SAF in government agencies and commissions, they
would not be the first ones to challenge a corporatist system of which they
themselves had become a part.
The activists were not loaded with the negotiators' and the experts' personal
experiences of becoming parts of a system of institutionalised interest
representation. Having their main base in the department of information,
and with some ties also to other functional units within the SAF bureaucracy,
the activists suggested that SAF play a more offensive political role, on the
basis of a clear-cut ideological identity, SAF's own problem definition, and
lobbyist and opinion-building activities.5
Thus, the activists were very uncomfortable with the Swedish model of
industrial relations. They argued that institutionalised cooperation with
other parties, such as LO, tended towards a shift in the SAF representatives'
loyalties, from principles of private business towards other parties such as
LO and public authorities. From this perspective the SAF representatives
were seen more as powerless hostages in a political game controlled by the
central LO, the activists' main enemy,6
than as any real counterweight to the
unions. From the activists' point of view the model was nothing but a euphemism
for continuing LO power.
310
The prospects for the activists to seriously threaten the model may seem
poor. After all, they constituted a minority group with relatively weak
anchorage in the leadership of SAF. At times, however, the underdog
influences politics,7
and in particular one would be wise not to forget the
skills usually associated with SAF's department of information in making
effective use of different kinds of persuasion and propaganda techniques. As
will become clear, these information-based techniques could be put to use
not only for the purposes of influencing public opinion and the politicians on
some political issue, but also in order to influence policy making within the
heterogeneous SAF.
The Policy-Making Process in SAF on the
Issue of Corporatism, 1982­19858
The Phase of Initiation
The seemingly robust support for corporatist arrangements was reflected in
the discussions among the leadership. The issue of corporatism that had been
discussed among the chief officials in 1979­1981 was a completely different
discussion from the activists' discussion about the evil of the Swedish model.
The chief officials had never called the model into question. It was rather the
tendency of being engaged in, and spending scarce organisational resources
on, issues that the interest organisations neither had a vital interest in nor
could claim to have any specific expert knowledge on that the chief officials
had reacted to and wanted to come to terms with (Artin & Arvidsson 1979;
Lindström 1979; Artin 1980; SAF's group of chief officials 1981).
In connection to that discussion, the chief officials were using the expression
`corporatism' as synonymous with institutionalised participation by
organised interests in the formulation of public policies outside their direct
areas of interests. According to that view, participation by the organisations
in decision-making bodies in the labour market sector was not seen as an
expression of `corporatism'. Those phenomena which the chief officials had
in mind when they dissociated themselves from corporatism were for example
the rather absurd representation of SAF and LO on the Treatment of
Offenders' board of directors and a couple of other government agencies on
the periphery of the labour market organisations' real spheres of interest
(Artin & Arvidsson 1979; Artin 1980; SAF's group of chief officials 1981).
Thus, the chief officials had identified a rationalisation problem for SAF
which had to be solved. That problem, furthermore, had consequences
beyond the issue of participation in public authorities ­ after all, scarce
organisational resources were also spent on participation in Royal Commissions
and the remiss process.9
311
The mission to deal with the task of trying to rationalise SAF's political
work was in 1982 given to Lennart Grafström, a person who formally had a
position on the mid level of the SAF bureaucracy as a kind of free resource,
utilised for the purposes of different kinds of investigations.
Grafström was, however, not only an experienced and efficient investigator;
he was also an activist, and occupied with the concerns of this group.In fact,
during his time as a leader of an internal SAF project, 1981­1982, Grafström
had became aware of the great power potential for LO connected to the system
of representation of labour market organisations in decision-making
bodies of powerful public authorities such as the National Labour Market
Board (AMS) and the National Board of Occupational Safety and Health
(ASS) (Grafström 1982a). Thus, in the autumn of 1982, when he was planning
a new investigation requested by the chief officials, Grafström set for
himself the visionary task of destroying the Swedish model, as far as corporatism
arrangements in AMS and ASS were concerned.10
From the point of view of Grafström, however, it would be necessary to
black out all controversial activist aspirations, and to present the investigation
plan in such a modest and appealing manner that it could receive the
requisite support from the chief officials.
Consequently, before the chief officials Grafström preferred to describe
the suggested two-year project on `the Influence of Organisations on Public
Authorities' in the following manner:
The analysis of corporatist tendencies ought to be deepening. To some extent attention to
the unions' influence on public authorities has been given within the project `Union-Led
Companies?' Within SAF an important debate on grounds of principle has started as far as
the role of private business is concerned.
There is now a need for an overall judgement of the influence upon public authorities by
the organisations. . . . The group ought also to make proposals as regards the future action of
SAF and in that connection among other things judge the possibilities of some kind of
`decorporatisation' in mutual understanding with the opposite parties. (Grafström 1982b, 6)11
Talking about `analysis' that `ought to be deepening' and mutual understanding
with the unions, Grafström made use of arguments carefully
designed for the deliberators/experts and the negotiators, respectively. And
having his project plan connected to an already existing `important debate on
grounds of principle' within SAF, thereby exploiting the ambiguity embedded
in the key term `corporatism', Grafström seemed to promise nothing
more than a logical follow-up of some earlier chief officials' discussions suggesting
changes on the margin of SAF's political work.
Thus, after a slight modification of the project name suggested by
Grafström ­ the suffix `Public Authorities' was abolished, indicating a
broader approach also encompassing the issues of participation in Royal
Commissions and the remiss process ­ the chief officials approved
Grafström's project idea. At the same time the chief officials agreed that
a straightforward, and generally well-reputed, deliberator12
would take
312
responsibility for the project as far as the chief officials were concerned
(SAMPOL 1983).
The Phase of Investigation
Thus, the issue of corporatism was supposed to be investigated by a group,
Influence of Organisations (OI), within SAF from 1983 to 1984. The group
was composed of some ten people, most of them mid-level officials employed
by SAF and the remaining ones brought in as representatives of some of the
SAF federations (SAMPOL 1983).
In terms of the SAF cultures I have introduced, OI at the beginning contained
two activists: Grafström, and Lars-Göran Redbrandt. Redbrandt had
been employed by SAF in the autumn of 1982, after serving six years as a
political expert in the chancellery of bourgeois governments. Grafström and
Redbrandt occupied positions in OI as project leader and secretary of investigation,
respectively (SAMPOL 1983).
Influence of Organisations also consisted of three experts brought in from
the SAF bureaucracy, and four representatives of SAF federations, which,
however, showed at most ­ an activist representing the federation of bakers
being a notable exception ­ a lukewarm interest in the investigation. Two of
the representatives had the image of being negotiators and one had a rather
vague image (SAMPOL 1983).
The multicultural composition of OI meant that different views were
expressed on the issue of corporatism during the investigation. From the
available documents it is clear that all three SAF experts expressed more or
less sceptical views about the activists' pledge for an abolition of interest
representation in public authorities' boards of directors within the main
area of SAF's activities (Grünewald 1983; 1984; Myrdal 1983; 1984; Regnell
1984).
All this scepticism notwithstanding, the activists eventually succeeded in
persuading the experts to support, or rather not to oppose, the recommendations
put forward in the final report from OI. The recommendations
showed a preference for a general abolition of interest representation in
public authorities' decision-making bodies, inclusive of AMS and ASS
(Influence of Organisations 1984d). How was this achieved?
During the time of investigation the activists in OI utilised their positions
in order to secure a privilege on the formulation of problems and control of
available solutions. Working in tandem, Grafström and Redbrandt carried
through most of the investigation's work on their own. They wrote all drafts
for discussions, including that of OI's final report, and the contributions of
the others were mainly restricted to comments on drafts written by Grafström
and Redbrandt on behalf of the group (Influence of Organisations
1984a; 1984b; 1984c; 1984d). The manner in which the two activists were
313
exercising power by controlling and biasing decision-relevant information is
worth considering in some detail.
In OI's final report (Influence of Organisations 1984d) the two activists
put forward three different models suggesting different routes for SAF's
future political work. The suggested proposal for OI implied a preference for
what was labelled a `strategy of developmenť, which was described as an
uncontroversial but at the same time innovative third alternative to the
`status quo' and `the radical strategy' (the latter was said to imply an immediate,
general and unilateral withdrawal by SAF from all kinds of formalised
participation within the political system).
The way in which the two activists packaged, labelled and connected different
proposals was very strategic. The idea of working for an abolition of
all corporatism arrangements in public authorities' boards of directors
(inclusive of AMS and ASS) was, for example, connected to and partly hidden
under the label of `the strategy of developmenť, a clever move if one
wished to play down the radical content of the anti-corporatist proposals
(Influence of Organisations 1984d).
The same purpose was served by the explicit and strong connection made,
in the final report, between the proposal of working for the abolition of interest
representation in public authorities' boards of directors and the idea of
working for the establishment of `permanent groups of reference'. The latter
idea suggested a preference for some kind of advisory bodies, composed of
representatives of organisations, connected to the public authorities with the
purpose of making continued formalised contacts with the public authorities
possible after the suggested abolition of corporatism in the public authorities'
boards of directors (Influence of Organisations 1984d).
By direct reference to this package of ideas, the proposal for a new SAF
policy on corporatism was described in the final report as a quite modest one
that `doesn't imply any dramatic change' (Influence of Organisations 1984d,
35).
In addition to this, the idea being symbolised by the advisory bodies organisations
as advisors communicating expert knowledge to the politicians
and the public authorities, thereby improving the intellectual basis of the
public policy-making process ­ of course was very much at the hearts of the
experts/deliberators, and the activists knew that.
The strategic choice of putting forward a proposal for a `strategy of
developmenť ­ at the expense of the activists' own preferred alternative,
`the radical strategy' ­ had, however, another implication of some importance.
With the aim of avoiding an irreconcilable conflict with the experts,
Grafström and Redbrandt temporarily gave up the idea of a unilateral withdrawal
by SAF from public authorities' boards of directors, such as those of
AMS and ASS (Influence of Organisations 1984c; 1984d). `Certain things
happen when you're writing a report; you have to master the art of what is
314
possible', Redbrandt states (interview, Redbrandt, 17 February 1999). At the
time, however, this didn't seem to be a huge sacrifice by the activists. Rather,
the ambition to have the anti-corporatist standpoint become an official SAF
policy would during the autumn of 1984 be the matter of most urgent importance.
So the activists gave priority to this.
Iťs also interesting to note that especially delicate issues about the investigation's
mandate, and internal anchoring of anti-corporatist ideas in SAF,
were never put on the agenda of OI. Instead, issues such as these were considered
in connection with special meetings aimed at drawing up plans for
further SAMPOL (SAF's Programme for Politics in Society) work,13
at
which Grafström and Redbrandt were the only attending members of OI.14
They were here surrounded by like-minded people, and in reality these
meetings became pure activist gatherings (SAMPOL 1984a; 1984b).
Thus, at the time when Grafström, in June 1984, described the anticorporatist
efforts as a `difficult internal process of change', which called for
`more powerful stimulation than hitherto', the activist programme group
had already expressed similar points of view (SAMPOL 1984c, 1). In notes
from a meeting held in March 1984, one can in fact read that `the group considered
it important to seek new means for internal organisational renewaľ
(SAMPOL 1984a, 2).
And ideas were hardly absent when the activists brought their minds
together. They would have to `keep on working with the mouth against
mouth methoď, and were planning `the production of real cases of successful
influence attempts and a concrete plan of abolition in some area' (SAMPOL
1984a, 2). They would also try to engage `externaľ actors, such as the Federation
of Industry and a few neoliberal opinion builders with connections to
private business (SAMPOL 1984a, 2).
The `mouth against mouth methoď was partly about trying ­ through personal
attendances ­ to win the experts and deliberators for anti-corporatist
ideas, partly about trying to get some more SAF activists engaged in the
project of dismantling the Swedish model.
`The production of real cases of successful attempts to influence' implied
an intention to search for concrete and real `cases' where SAF's opinionbuilding
and lobbying activities had turned out to be successful in terms of
influence upon public policy (note that all kinds of failures in this respect,
which could complicate the overall picture and, consequently, the anticorporatist
efforts, were to be excluded). This, of course, was about convincing
potential sceptics about the efficiency of informal, non-institutionalised
methods of political work in terms of influence upon public policy. In the
final report from OI, 11 (!) such `real cases' were introduced in an appendix
(Influence of Organisations 1984d).
It was, further, considered important to carry through quite soon a
unilateral SAF withdrawal from at least some public authorities' boards of
315
directors which were considered peripheral in relation to SAF's main areas
of activities. The underlying idea of the activists with respect to that move
was not that such withdrawal ought to serve as a basis for evaluation, and
that such experiences ought to be discussed in a deliberative discussion
within SAF about the desirability of withdrawal from the public authorities'
boards of directors in SAF's main fields of activity. This was rather intended
as `a clever move' with the aim of giving sceptics towards anti-corporatist
ideas `some little sign of what probably would happen' (quoted from interview,
Redbrandt; Grafström 1985).
During the time when OI was working, regular reports about the work
were presented before the chief officials. Even if the group of chief officials
did not have any formal decision-making power as a collegial body, it was,
of course, of great importance for the work of OI how these reports were
dealt with by the chief officials.
One serious problem for the mid-level activists to deal with in this context
was that of the very sharp cultural differences between the activists and the
negotiators within SAF. Having been lucky enough to avoid having SAF
negotiators in OI, they sooner or later had to face the almost impossible task
of convincing SAF's chief negotiator15
­ a powerful man who was the frontfigure
among the negotiators and also the managing director's deputy about
the merits of anti-corporatist proposals.
The mid-level activists' way of dealing with this problem was to make sure
that as little information about what was really going on in OI as possible was
communicated to the chief negotiator. Thus, when preparing the presentations
to be done before the collegial body of the group of chief officials, the
aspirations of bringing an end to the system of interest representation in
AMS and ASS were blacked out.
At the same time, successful efforts were made to anchor these ideas
among the deliberators among the chief officials, by putting forward arguments
of principle, coupled with the suggestion of advisory bodies. In early
1984, the deliberator responsible for the SAMPOL work among the chief
officials changed his view on interest representation in AMS's and ASS's
decision-making bodies (Unge 1984). Since he was a person known for
his good and thoughtful judgements, this also paved the way for a rapid
change of view among other individuals in the collegial group of chief
officials, among them the managing director (Grafström 1984; Ljunggren
1984).
Thus, at the time ­ in December 1984 ­ when the final report was on the
collegial group of chief officials' meeting table, the anti-corporatist proposals
were already anchored among most of the chief officials. The chief negotiator,
however, was quite upset, stating his opinion that `making anti-corporatism
now a superior concept in SAF's activities has been announced too late', and
defending the status quo in AMS and ASS in regard to corporatism (SAF's
316
group of chief officials 1984, 3). But it was the chief negotiator himself who
was too late. The deliberators were now out of reach. The chief negotiator
had been defeated by the mid-level activists (SAF's group of chief officials
1984).
The Phase of Decision Making
The Swedish Employers' Confederation's board of directors had not been a
driving force in the process of initiating and investigating the issue of corporatism.
In regard to the work of investigation, this is hardly surprising. At this
time it was not supposed to be the boarďs task. Tasks of investigation were
instead supposed to be carried out by the SAF bureaucracy. But making
decisions on fundamental issues of policy, such as that of SAF's view on
interest representation in AMS and ASS, was definitely something to be
decided by SAF's board of directors. How did the board deal with this task?
The remarkable answer is that the issue of corporatism was never discussed
on the board, although the issue was actually on the meeting agenda
on three occasions before the decision on a fundamental change of SAF
policy finally was taken.
A first oral report on OI given by Grafström in January 1984 was followed
by a suggestion from the chairman of SAF16
to postpone the issue. That
settled the matter at this stage (SAF's board of directors 1984).
The next time corporatism appeared on the boarďs agenda was in March
1985, when the final report from OI was presented by SAF's managing dir-
ector.17
The final report, which was distributed among the members of the
board before the meeting, proposed an unambiguous anti-corporatist view
on the issue of the composition of public authorities' boards of directors,
something that was also highlighted in the oral presentation made by the
managing director. Discussion was, however, still absent (SAF's board of
directors 1985a).18
One member of the board19
did express a view at the March meeting, stating
his opinion that `an important and great work had been carried ouť by
OI (SAF's board of directors 1985a, 10). `I consider the investigation so
important that it deserves a discussion at a future time,' he concluded,
thereby in effect arguing for a new non-decision (SAF's board of directors
1985a, 10). The Swedish Employers' Confederation's board of directors
followed that recommendation, deciding `an extra meeting of the board
of directors' should `take place on June 14 . . . for a deepened discussion
about the project "Influence of Organisations" ' (SAF's board of directors
1985a, 11).
Such an extra meeting of the board was, however, never to take place, and
it was not until November 1985 that the issue of corporatism was considered
again by this body. On this occasion, the board was supposed to communicate
317
its view on a draft, written by Grafström, expressing SAF's public view on
the final report from a parliamentary ad hoc commission.20
Having been set
up by the Social Democratic government in 1983 with the purpose of investigating
the forms of governance of, and by, public authorities, among other
things the commission proposed a general abolition of interest representation
within the public authorities' boards of directors.
In the draft the proposed SAF position was formulated as `SAF shares the
point of view of the . . . committee that interest organisations ought not to be
represented in the public authorities' boards of directors' (SAF's board of
directors 1985b, Appendix 3:1). The Swedish Employers' Confederation's
board of directors could hardly have formulated an anti-corporatist standpoint
on the issue of representation in the public authorities' boards of directors
much clearer than this, and, with formulations apparently marking
a fundamental change of SAF's standpoint on an important issue of policy,
one would now expect a thorough discussion and debate within the board of
directors.
But the expected discussion never came. Immediately after Grafström's
oral presentation, the board took a unanimous decision implying approval of
Grafstrom's draft of SAF's view without any sign of reservation (SAF's
board of directors 1985b). The decision on principles, marking a willingness
to work for a general abolition of corporatism in public authorities' boards
of directors, was in fact a decision taken without any preceding discussion
within SAF's board of directors. How come?
Calling for non-decisions when the issue was on the meeting table of
SAF's board of directors, and being successful in imposing their will on the
board in that respect, two members of the board certainly contributed to the
course of events.
Having said that, I think, however, that the most critical questions still
need to be answered if one wants to understand why a fundamental change
of SAF policy was decided without any discussion within SAF's board of
directors. Why did the scheduled extra meeting in June 1985, when the final
report from OI was to be discussed, never take place? How come there was
no discussion at the decisive, policy-changing meeting in November 1985?
In early February 1985, the managing director had set up a particular
Implementation Group within the SAF bureaucracy with the purpose of further
elaboration of the intentions in the final report from OI. Grafström and
Redbrandt were included in that group, and contributed to the fact that from
the start the group acted strategically in its relations to SAF's board of directors
(Unge et al. 1985).
Even before a meeting to be held in late April 1985, the secretary of the
group, Redbrandt, had made the issue an individual point of discussion on
the Implementation Group's agenda (Redbrandt 1985a). And from documents
recording that meeting one can read that the group also instructed the
318
managing director to utilise the legal means at his disposal to cancel the extra
meeting of SAF's board of directors which was supposed to take place in
June (Implementation Group 1985).
Furthermore, before the November meeting of the board of directors in
1985, the officials had, in the written basis for discussion, presented a quite
peculiar interpretation of the previous March meeting of the board of directors.
`[G]eneral support for the conclusions and propositions' being codified
in the final report from OI was claimed to have existed at the latter meeting
(Ljunggren 1985, 1). Such an interpretation of the March meeting seems
wrong or at least highly questionable. The fact is that only one member of
the board had expressed support for the report. Otherwise, the report was
dealt with indecisively.
The interpretation put forward by the officials ­ from the contemporary
records of the work within the Implementation Group one could identify
Redbrandt as the person planning this move (Unge 1985a; 1985b; Redbrandt
1985b) ­ implied that the members of the board were told that there already
existed an anti-corporatist policy which the November meeting only had to
acknowledge and codify, rather than face the task of making a brand new
decision on a change of SAF policy. Further, this interpretation had already
had some practical consequences. Within the SAF bureaucracy, extensive
work had already been carried out by the Implementation Group on further
development and implementation of the intentions in the final report from
OI. This was also something that was communicated in writing to the board
before the November meeting (Unge 1985c).
In addition to that, the board was also told by Grafström, at his oral presentation
at the November meeting, that even if corporatism in decisionmaking
bodies of public authorities were to be abolished, SAF would still,
and for the foreseeable future, advocate formal pathways by which the
organisation should exercise influence on the public authorities (SAF's
board of directors 1985b).
Small wonder, then, that discussion was completely absent at the decisive,
policy-changing meeting of SAF's board of directors in November 1985.
The Phase of Implementation
From the contemporary sources, constituting records of the investigations
into SAF's political work, it is clear that the like-minded mid-level activists
Grafström and Redbrandt were as early as in the winter of 1983/1984 establishing
contacts with important elements of the political system in Sweden.
The main object of attention was the parliamentary ad hoc commission
instructed in 1983 by the Social Democratic government to investigate and
make proposals on reforms of the forms of governance of, and leadership
models within, public authorities (SAMPOL 1984b; 1984c).
319
The available sources tell us that Redbrandt had frequent contacts with
the representative of private enterprise in the commission, where formally
no representatives of interest groups were to have seats, and also that direct
contacts were established by Redbrandt and Grafström with the three representatives
of the non-socialist political parties within the commission
(SAMPOL 1984b; 1984c). In fact, Redbrandt was already familiar with these
party politicians from his six-year period as a political expert within a department
of the Swedish government.
What kind of purposes did these informal contacts ­ between people from
SAF and members of the parliamentary commission ­ serve, from an SAF
perspective?
The answer depends on through which SAF cultural lens one considers
these contacts. In the mind of a straightforward deliberator ­ in fact, the only
person among SAF's leadership who was lucky enough to be informed
that contacts were established between SAF and the commission was a
straightforward deliberator21
­ the contacts would serve the purpose of
getting some more information about the issue of corporatism and, consequently,
broadening the basis and the quality of the work of the SAF
investigation (Unge 1983; SAF's group of chief officials 1984). They would,
consequently, strengthen the deliberative character of policy making within
SAF.
The two mid-level activists, however, didn't care much for deliberation.
They had their own great mission: to bring down a corner pillar of the Swedish
model. Consequently, they were in the first place interested in influencing
public policy rather than collecting information in order to broaden the basis
of SAF's own investigation of the issue of corporatism.
These lobbying activities seem remarkable in two ways. First: in its final
report, the parliamentary commission actually put forward proposals in
which the requests of the mid-level activists were wholly satisfied, not least
on the issue of participation by the organisations in decision-making bodies
of the public authorities. Second: two officials in the mid level of the SAF
hierarchy obviously tried to anchor anti-corporatist ideas in central parts of
the political system of Sweden before SAF's board of directors, the antagonistic
chief negotiator and some others among the chief officials had been
given the opportunity to declare, and argue for, their positions on the point
at issue.
The two mid-level activists seem, further, to have been the driving forces
when the above-mentioned Implementation Group was set up in February
1985 within the SAF bureaucracy for further development of the intentions
in the final report from OI (Influence of Organisations 1984d). The remarkable
thing about this new constellation ­ in which Grafström and Redbrandt
had taken seats ­ was that it was set up even before the final report of OI had
been presented to SAF's board of directors.
320
Conclusions
In 1985 SAF's board of directors decided to recommend the abolition of
formal interest representation in decision-making bodies of government
agencies ­ a serious stroke against a corner pillar in the Swedish model of
corporatist institutions in industrial relations. Using the method of process
tracing, the aim of this article has been to test the standard hypothesis among
industrial relations analysts, which stresses the leading role of the organisation's
leadership in this process of policy change.
The results clearly indicate that a few individuals in the mid level of the
SAF hierarchy were the most powerful actors and proponents in the policymaking
process, succeeding in getting their own anti-corporatist ideas
anchored into the leadership of SAF and, consequently, raised to official
SAF policy.
At a first glance, the results of the mid-level officials' efforts in terms of
influence on SAF policy seem rather puzzling. After all, they had to work
within an overwhelmingly `model-friendly' organisational environment,
where ideas of mutual understanding with opposite parties (the central
organisations of unions) and of organisations as `deliberators' within formal,
corporatist institutions were widespread. And foremost: the model had a
strong hold on the leadership within SAF, which had never called into question
interest representation in public authorities' boards of directors ­ within
the labour market and working life spheres ­ before mid-level officials put
the issue upon the internal SAF agenda.
A closer look at the policy-making process within SAF, however, reveals
that the mid-level officials were controlling a dangerous weapon that could
compensate for the lack of formal positions within the organisation: the flow
of decision-relevant information. In fact, the mid-level officials pushing for
anti-corporatist reforms belonged to a specific activist minority culture
within SAF, with its main base in SAF's department of information; a
department well and widely known for its skills in making effective use of
different persuasion and propaganda techniques.
There are plenty of examples in the article on how the mid-level activists
were deliberately biasing and controlling decision-relevant information in
order to influence SAF policy. I have also shown how they succeeded in their
efforts to steer the huge SAF ship in an anti-corporatist direction, at all the
different stages ­ the phases of initiation, investigation, decision and implementation
­ in the policy-making process within the organisation.
Thus, from an organisational level the policy-making process seems quite
irrational. Efforts were made to implement a decision not yet taken (for example
when the mid-level activists tried to influence an ad hoc parliamentary
commission without an SAF mandate to lean against). And when the decision
on a change of SAF policy finally was taken, it was a decision without
321
any preceding discussion within SAF's board of directors (the acknowledged
policy-making body within SAF). A great deal of investigative work was
done on the issue of interest representation in public authorities' boards of
directors, but this work was mainly done by people in SAF who from the
beginning had convinced themselves what kind of recommendations
the investigation ought to present. The list of examples could be made
longer.
There are general insights into organisational decision making to be learnt
from the case study of policy change in SAF. Loosely coupled units and
the existence of subgroups within organisations which have interests and
priorities other than those of the formal leadership are common phenomena
in organisational life. And if individuals of those subgroups also happen to
possess some skill in the art of strategic manipulation of information, they
are potentially powerful actors in organisations whenever they have the
opportunity to undertake investigations and other kinds of preparatory work
on behalf of the organisation.
The implications for modern policy analysis are obvious. One cannot take
for granted that the organisational leadership plays as dominant a role in
policy-making processes as is suggested in simplistic rationalistic models
of policy making. On the other hand it would be equally misleading to
assume a priori that policy making resembles some garbage can process
or follows some logic of appropriateness. Policy-making processes are
most of the time steered by certain people. These people, however, are often
mid-level officials claiming to be acknowledged and legitimate agents of
change by their cover names of `investigators' or neutral `experts'.
ACKNOWLEDGEMENTS
The article was written with financial aid from the National Institute for Working Life,
Sweden. I would also like to thank PerOla Öberg and Kaare Vernby at the Department of
Government, Uppsala, and three anonymous referees for valuable comments on earlier
versions of the article.
NOTES
1. The implicit view of a leadership-driven process of policy change is to be found in those
studies that treat SAF as a homogeneous actor and which, consequently, avoid issues
about power in the intra-organisational process of policy change (Pestoff 1995; Rothstein
& Bergström 1999; cf. Kjellberg 1992; Lewin 1992; Crouch 1994; Öberg 1997).
About policy making in general in SAF and the supposed central role of the leadership,
see Bresky et al. (1981), De Geer (1986) and Jerneck (1986) (cf. Söderpalm 1976; 1980;
Hansson 1984; Hansson & Lodenius 1988; Rothstein 1992; Boréus 1994; Stenls
1998).
2. About the formal structure and the statutes of SAF see De Geer (1986).
3. The distinction between the four phases should be seen as a way to present the empirical
data, trying to make sense of the process of policy making in SAF by capturing its
322
main characteristics. It is regarded as an empirical question to what extent the process
really follows the sequential logic suggested above. (Cf. Premfors 1989; Lindblom &
Woodhouse 1993.)
4. The cultures I'll describe represent different views on the appropriate role of SAF in
the political system of Sweden. These views are, further, strongly correlated with the
SAF individuals' main functional tasks in the organisation. Thus, the cultural conflict
I'll describe corresponds to one of the most well-known and acknowledged types of
cultural conflicts in organisations. The division into three cultures and the description
of these cultures is my own device, but I have also been inspired by others, foremostly
De Geer (1989). The classification of an SAF individual into one of these cultures is
done by me on the basis of the sources I refer to in Johansson (2000, 153­79).
5. A more extensive account of these activist discussions is to be found in Johansson
(2000).
6. See Johansson (2000) for a discussion about central LO as the main enemy in the eyes
of the SAF activists, and about the related issue of wage-earner funds.
7. Cf. Kaarbo (1998).
8. The study is mainly based on studies in the archives of SAF, to which I have been given
unlimited access. In respect to this, I have scrutinised all protocols from meetings with
SAF's board of directors and meetings with SAF's group of chief officials from 1979 up
to and including 1985, all available records from the investigation in SAMPOL, 1983­
1984, and the follow-up activities of the Implementation Group, and also some other
SAF internal documents. Some additional information has been collected from other
types of sources, which have also served as cross-checking devices in relation to inferences
on the basis of documents from the archives of SAF. A systematic investigation
has been accomplished with regard to all protocols from the Federation of Engineering
Industries (VF's) board of directors from 1980 up to and including 1985 (some arguments
for this focus on VF of all possible SAF federations are presented below in note
10), all issues of the membership journal of SAF (SAF-tidningen) from 1980 up to and
including 1985, and the official publications from the SAF congresses (SAF-kongressen)
that were held in 1980 and 1984. In addition to this I have also used collections of newspaper
articles, debate books, semi-structured interviews, etc. in order to get a more comprehensive
view of the subject. A more extensive account of the sources I draw on in
my empirical analysis, and my methodological considerations in the process of selecting
and, not least important, interpreting these sources, may be found in Johansson (2000).
9. In the political process of Sweden, Royal Commissions are appointed to conduct
research and make recommendations on laws. The remiss process, on the other hand,
allows all interested parties to express their views on the Royal Commission's recommendations,
and these views are to be considered by the government before it puts forward
its proposals on laws in the Riksdag.
10. A more extensive discussion about Grafström's role during the phase of initiation may
be found in Johansson (2000), where I also discuss the role of SAF federations such as
the Federation of Engineering Industries (VF), an actor that according to industrial
relations analysts has been a very important agent of change in the contemporary twin
issue of decentralisation of the system of wage bargaining (Elvander 1988; Swenson &
Pontusson 2000), and that, consequently, could be expected to have been an agent of
change also in the issue of corporatism in AMS and ASS. Having analysed all documents
from meetings with VF's board of directors from 1980 up to and including 1985
(VF's board of directors 1980­1985; VF's board of directors working committee 1980­
1985), and also writings from and interviews with the big man in VF, Pehr G. Gyllenhammar
(interview, Gyllenhammar, November 1983; Gyllenhammar 1991), I have,
however, found no evidence at all supporting the idea of VF as an important actor in
the policy-making process of SAF in the issue of corporatism in 1982­1985. This conclusion
is, further, wholly consistent with the picture given by the other types of sources
­ internal SAF documents, public sources, semi-structured interviews, etc. ­ I have analysed.
See Johansson (2000).
11. All Swedish sources I quote in this article have been translated by me into English.
12. Torkel Unge.
323
13. The programme was to be carried out during a two-year period, 1983­1984, and consisted
of three different SAF internal investigations: `Tendencies of Development for
Unions in Western Europe', `Control of Firms' and `Influence of Organisations'. See
Johansson (2000).
14. The programme group was composed of Grafström (who was the leader of all three
investigation groups) and the three respective secretaries of the investigations, who also
happened to be activists. The deliberator, who was supposed to take responsibility for
the programme work among the chief officials, didn't attend these meetings (SAMPOL
1984a; 1984b). He seems to have preferred a delegating style of leadership on these
issues. For further information see Johansson (2000).
15. Lars-Gunnar Albge.
16. Curt Nicolin.
17. Olof Ljunggren.
18. Nicolin had now been succeeded by Claes-Ulrik Winberg as chairman of SAF.
19. Pehr G. Gyllenhammar.
20. Verksledningskommittén.
21. The other chief officials were not informed until the final report from OI was on the
agenda of the collegial group of chief officials, in December 1984. The antagonistic chief
negotiator was very upset about that (SAF's group of chief officials 1984).
REFERENCES
Artin, S. 1980. `Idéskiss för projekť, Draft, 9 September, P00202:3, Archives at SAF.
Artin, S. & Arvidsson, L.-E. 1979. `SAFs representation i samhällsorgan', Investigation paper,
30 July, P063:19a, Archives at SAF.
Blyth, M. 2001. `The Transformation of the Swedish Model. Economic Ideas, Distributional
Conflict and Institutional Change', World Politics 54, 1­26.
Boréus, K. 1994. Högervg. Nyliberalism och kampen om sprket i svensk offentlig debatt 1969­
1989. Stockholm: Tiden.
Bresky, T., Scherman, J. & Schmid, I. 1981. Med SAF vid rodret. Granskning av en kamporganisation.
Stockholm: Liber Förlag.
Crouch, C. 1994. `Beyond Corporatism: The Impact of Company Strategy', in Hyman, R. &
Ferner, A., eds, New Frontiers in European Industrial Relations. Oxford: Blackwell.
Cyert, R. M. & March, J. G. 1963. A Behavioral Theory of the Firm. Englewood Cliffs, NJ:
Prentice-Hall.
De Geer, H. 1986. SAF i förhandlingar. Svenska Arbetsgivareföreningen och dess förhandlingsrelationer
till LO och tjänstemannaorganisationerna 1930­1970. Stockholm: Svenska
Arbetsgivareföreningen.
De Geer, H. 1989. I vänstervind och högervg. SAF under 1970-talet. Stockholm: Allmänna
Förlaget.
De Geer, H. 1992. Arbetsgivarna. SAF i tio decennier. Stockholm: Svenska
Arbetsgivareföreningen.
Elvander, N. 1988. Den svenska modellen. Löneförhandlingar och inkomstpolitik 1982­1986.
Stockholm: Allmänna Förlaget.
Grafström, L. 1982a. `Facklig företagsledning?', in Bratt, C. & Johansson, M., eds, Facklig
företagsledning? Vägval inför 90-talet. Stockholm: Svenska Arbetsgivareföreningen.
Grafström, L. 1982b. `LEG-program: Samhällspolitik och företagsstyrning. Förslag', Draft, 15
December, P00202:1, Archives at SAF.
Grafström, L. 1984. `Betr.: Makro-gruppen disk m A 1984-02-06', Message, 8 Feburary,
P00202:1, Archives at SAF.
Grafström, L. 1985. `Samhällspolitiskt praktikfall. Arbetsmarknadsparterna skall ej medverka i
myndighetsbeslut; pverkansplan oktober 1985', Message, 10 October, Archives at SAF.
Grünewald, B. 1983. `Korporativism inom utbildningsomrdeť, Paper, 25 November, P00202:1,
Archives at SAF.
Grünewald, B. 1984. `Beräknat resursbehov för antikorporativistisk utbildningspolitik jämte
kommentarer', Paper, 4 April, P00202:1, Archives at SAF.
324
Gyllenhammar, P. G. 1991. Även med känsla. Stockholm: Bonnier Alba Bokförlag.
Hansson, S. O. 1984. SAF i politiken. En dokumentation av näringslivsorganisationernas opinionsbildning.
Stockholm: Tidens förlag.
Hansson, S. O. & Lodenius, A.-L. 1988. Operation högervridning. Stockholm: Tidens förlag.
Implementation Group. 1985. `SAMPOL-anteckningar 3', Notes from the meeting on 29 April,
Archives at SAF.
Influence of Organisations. 1984a. `Principer för SAFs framtida samhällspolitiska arbete', Draft,
14 February, P00202:1, Archives at SAF.
Influence of Organisations. 1984b. `Vägledning för SAFs framtida samhällspolitiska arbete',
Half-time report, April, P00202:1, Archives at SAF.
Influence of Organisations. 1984c. `Synpunkter p SAFs framtida samhällspolitiska arbete',
Paper without a note of date, P00202:1, Archives at SAF.
Influence of Organisations. 1984d. `Strategi, Aktion, Framgng ­ om SAFs samhällspolitiska
arbete', Final report, November, Archives at SAF.
Jerneck, M. 1986. SAFs framtidssyn: förutsägelser, mlsättningar och dilemman. En studie av
programarbete och planeringsverksamhet inom Svenska Arbetsgivareföreningens kansli.
Stockholm: Svenska Arbetsgivareföreningen.
Johansson, J. 2000. SAF och den svenska modellen. En studie av uppbrottet frn
förvaltningskorporatismen 1982­91. Uppsala: Acta Universitatis Upsaliensis.
Kaarbo, J. 1998. `Power Politics in Foreign Policy: The Influence of Bureaucratic Minorities',
European Journal of International Relations 4, 67­97.
Kjellberg, A. 1992. `Sweden: Can the Model Survive?', in Ferner, A. & Hyman, R., eds, Industrial
Relations in the New Europe. Oxford: Blackwell.
Lewin, L. 1992. Samhället och de organiserade intressena. Stockholm: Norstedts.
Lindblom, C. E. & Woodhouse, E. J. 1993 [1968]. The Policy-Making Process, 3rd edn. Englewood
Cliffs, NJ: Prentice Hall.
Lindström, G. 1979. `Frgan om arbetsgivareföreningen bör vara representerad i vissa offentliga
och enskilda organ', Paper, 14 September, P063:19b, Archives at SAF.
Ljunggren, O. 1984. `De västeuropeiska modellerna ­ svensk samhällsorganisation i en spegeľ,
Talk, 3 October, P00202:1, Archives at SAF.
Ljunggren, O. 1985. `SAFs uppgifter ­ inriktning och avgränsning', Letter, 24 June 1985,
Archives at SAF.
Myrdal, H.-G. 1983. `SAFs policypapper beträffande organisationernas inflytande', Message, 4
October, P00202:1, Archives at SAF.
Myrdal, H.-G. 1984. `Särskilt yttrande', Comment on the final report from OI, 6 December,
Archives at SAF.
Pestoff, V. A. 1995. `Towards a New Swedish Model of Collective Bargaining and Politics', in
Crouch, C. & Traxler, F., eds, Organized Industrial Relations in Europe: What Future? Aldershot:
Avebury.
Pontusson, J. 2002. `Joakim Johansson: SAF och den svenska modellen. En studie av uppbrottet
frn förvaltningskorporatismen 1982­91. Anm.', Book review, Statsvetenskaplig Tidskrift 104,
172­78.
Premfors, R. 1989. Policyanalys. Kunskap, praktik och etik i offentlig verksamhet. Lund:
Studentlitteratur.
Redbrandt, L.-G. 1985a. `Förslag till dagordning för SAMPOL den 29 april 1985', Paper without
a note of date, Archives at SAF.
Redbrandt, L.-G. 1985b. `SAMPOL 1985-06-05. Förslag till föredragningslista', Paper without
a note of date, Archives at SAF.
Regnell, M. 1984. `Synpunkter p korporativismen inom arbetsmarknadsförvaltningen
närmast arbetsmarknadsverkeť, Draft for discussion, 9 January, P00202:1, Archives at
SAF.
Rothstein, B. 1992. Den korporativa staten. Intresseorganisationer och statsförvaltning i svensk
politik. Stockholm: Norstedts.
Rothstein, B. & Bergström, J. 1999. Korporatismens fall och den svenska modellens kris. Stockholm:
SNS Förlag.
SAF's board of directors. 1984. `Protokoll, fört vid sammanträde med Svenska Arbetsgivareföreningens
styrelse torsdagen den 19 januari 1984', Archives at SAF.
325
SAF's board of directors. 1985a. `Protokoll fört vid sammanträde med Svenska Arbetsgivareföreningens
styrelse torsdagen den 21 mars 1985', Archives at SAF.
SAF's board of directors. 1985b. `Protokoll fört vid sammanträde med Svenska Arbetsgivareföreningens
styrelse torsdagen den 14 och fredagen den 15 november 1985', Archives at SAF.
SAF's group of chief officials. 1981. `Vad skall SAF syssla med? Fas A', Internal report, 3
December, Archives at SAF.
SAF's group of chief officials. 1984. `Minnesanteckningar frn Ledningsgruppens sammanträde
mndagen den 17 december 1984', Notes from a meeting, minutes dated 9 January 1985,
Archives at SAF.
SAMPOL. 1983. `Samhällspolitik och företagsstyrning ­ ett nytt SAF-program', Paper, 13 January,
P00202:1, Archives at SAF.
SAMPOL. 1984a. `Minnesanteckning. Frn: Sammanträde 1984-03-08', Notes from a meeting,
12 March, P00202:1, Archives at SAF.
SAMPOL. 1984b. `SAMPOL ­ planeringssammanträde 1984-04-16', Notes from a meeting,
25 April, P00202:1, Archives at SAF.
SAMPOL. 1984c. `Samhällspolitiska programmet. Redovisning i LEG, 1984-06-12', Paper,
7 June, P00202:1, Archives at SAF.
Stenls, N. 1998. Den inre kretsen. Den svenska ekonomiska elitens inflytande över partipolitik
och opinionsbildning 1940­1949. Lund: Arkiv förlag.
Swenson, P. & Pontusson, J. 2000. `The Swedish Employer Offensive against Central Wage Bargaining',
in Iversen, T., Pontusson, J. & Soscice, D., eds, Unions, Employers and Central
Banks. Macroeconomic Coordination and Institutional Change in Social Market Economies.
Cambridge: Cambridge University Press.
Söderpalm, S. A. 1976. Direktörsklubben. Storindustrin i svensk politik under 1930- och 40-talen.
Stockholm: Zenit/Rabén & Sjögren.
Söderpalm, S. A. 1980. Arbetsgivarna och Saltsjöbadspolitiken. En historisk studie i samarbetet
p svensk arbetsmarknad. Stockholm: Svenska Arbetsgivareföreningen.
Unge, T. 1983. Letter without a heading, sent to Rune Molin, LO, 28 September, P00202:1,
Archives at SAF.
Unge, T. 1984. Message without a heading, sent to representatives in national public authorities
nominated by SAF, 22 February, P00202:1, Archives at SAF.
Unge, T. 1985a. `Angende den fortsatta behandlingen i styrelsen av Sampol-programmeť,
Message, 21 May, Archives at SAF.
Unge, T. 1985b. `Skiss till PM att tillställas styrelsen inför dess fortsatta behandling av rapporten
"Strategi, Aktion, Framgng" vid sammanträdet p Yxtaholm den 14­15 november 1985',
Draft, 21 May, Archives at SAF.
Unge, T. 1985c. `SAFs uppgifter ­ inriktning och avgränsning', Paper, 24 June, Archives at SAF.
Unge, T., Grafström, L. & Redbrandt, L.-G. 1985. `Genomförandegruppen', Paper, 8 February,
Archives at SAF.
VF's board of directors. 1980­1985. All protocols from the meetings with the board of directors
from 1980 up to and including 1985, Archives at the Federation of Engineering Industries.
VF's board of directors working committee. 1980­1985. All protocols from the meetings with
the board of directors' working committee from 1980 up to and including 1985, Archives at
the Federation of Engineering Industries.
Öberg, P. O. 1997. `Medborgarnas inflytande och särintressenas makt. Korporatism och lobbying
i statsförvaltningen', Rapport till förvaltningspolitiska kommissionen. Politiska institutioner
och strategiskt agerande, 17, Department of Government, University of Uppsala.