Political Exchange and Bargaining Reform in Italy and Spain

Abstract

Wage bargaining structures in Italy and Spain changed significantly in the 1990s. This is usually seen as an employer-led response to exogenous pressures such as the Economic and Monetary Union (EMU). This article shows that while EMU acted as a catalyst for negotiated adjustments, changes in wage bargaining are better explained through factors endogenous to national systems, in particular union strategies and interactions in the policy-making arena. By means of policy concertation and political exchange, unions have shaped institutional change in collective bargaining so as to avoid a disorganized decentralization of labour relations.

Full text

Óscar Molina Romo
European University Institute, Badia Fiesolana, ITALY
Political Exchange and Bargaining
Reform in Italy and Spain
ABSTRACT
▪
Wage bargaining structures in Italy and Spain changed
significantly in the 1990s. This is usually seen as an employer-led response to
exogenous pressures such as the Economic and Monetary Union (EMU). This
article shows that while EMU acted as a catalyst for negotiated adjustments,
changes in wage bargaining are better explained through factors endogenous
to national systems, in particular union strategies and interactions in the
policy-making arena. By means of policy concertation and political exchange,
unions have shaped institutional change in collective bargaining so as to
avoid a disorganized decentralization of labour relations.
Wage Bargaining under Stress: Convergence or
Divergence?
The last decade has witnessed an extensive reform of collective bargain-
ing structures in many EU countries (Pochet, 1999). These reforms have
been triggered by the interaction between a series of pressures: exogen-
ous (the need to comply with EMU) and endogenous (the impact of
changing labour markets). Interpretation and projection have presented
two scenarios (Regini, 2000): the first, employer-led view predicts deregu-
lation and decentralization through the substitution of collective by indi-
vidual (market-driven) regulation, a process which trade union
movements are now too weak to resist. The outcome, according to this
scenario, would be wage moderation triggered by the decentralization of
wage determination and the gradual elimination of barriers to free indi-
vidual bargaining between employers and workers, thus leading to wider
wage differentials. A second scenario (Pérez, 2002; Soskice, 1990)
discounts the probability of across-the-board deregulation and disorgan-
ized decentralization; more probable is a process of growing wage
coordination, with beneficial effects on competitiveness. In this view,
change derives from inter-associational consensus supported by govern-
ment rather than pure employer initiatives. Both scenarios coincide to
European Journal of Industrial Relations
© 2005 SAGE (London, Thousand Oaks, CA and New Delhi)
Volume 11 Number 1 pp 7–26 www.sagepublications.com
DOI: 10.1177/0959680105050397
02 romo (ds) 21/1/05 10:27 am Page 7
peer-00570929, version 1 - 1 Mar 2011
Author manuscript, published in "European Journal of Industrial Relations 11, 1 (2005)
7-26"
DOI : 10.1177/0959680105050397

give employers and/or governments a leading role in the process of
change, hence predicting convergence in wage bargaining institutions and
outcomes, reflecting the weakness of unions and the disciplining role of
EMU.
However, detailed studies of changes in industrial relations and
bargaining systems do not seem to indicate convergence. Instead, they
point towards persisting differences (EIRO, 1999; Kauppinen, 1998).
One possible explanation derives from the neo-institutionalist ‘varieties
of capitalism’ approach (Hall and Soskice, 2001): different social modes
of production and economic performance are embedded in complex
systemic configurations of social and economic institutions which are
functionally interdependent and hence resistant to change. When insti-
tutional change occurs it therefore usually reflects powerful exogenous
pressures and is costly, path dependent and (because it is company-
centred) employer-led.
A second approach stresses the role of actors’ strategic choices
(Kochan et al., 1986): cross-national differences are traced back to
particular configurations of power relations among corporate social
actors with capacity to influence national public policy-making. Insti-
tutional change results from actors’ interactions and new institutional
equilibria reflect changes in their balance of power, strategies and forms
of interaction. A third variant is provided by actor-centred institutional
accounts, which focus on rational actor behaviour within the limits
imposed by the ‘system’, or the institutional framework that determines
not only feasible adjustment paths but actors’ strategies (Scharpf, 1997).
According to this approach (Wailes et al., 2003), strategies are shaped not
only by an individual actor’s calculations, nor the dominant traits of the
system (coordinated or disorganized), but also the political, economic
and institutional context. Contrary to the ‘varieties of capitalism’
argument, this view can explain differences in the direction of insti-
tutional change between countries with similar ‘social systems of produc-
tion’.
These approaches provide different accounts of the source of change
in collective bargaining systems, their driving mechanisms and direction.
This article offers an alternative explanation through a comparison of
Italy and Spain: countries experiencing major structural change in
bargaining systems. Given that both collective bargaining systems
initially had similar structures, explanations of differences need to
address factors other than purely institutional. The analysis below
follows actor-centred institutionalism, but departs from existing litera-
ture in three significant aspects. First, from the point of view of the
dependent variable, it analyses changes in bargaining systems in terms of
their articulation. This is because this concept captures both vertical
(centralization vs. decentralization) and horizontal (coordination vs.
European Journal of Industrial Relations 11(1)
8
02 romo (ds) 21/1/05 10:27 am Page 8
peer-00570929, version 1 - 1 Mar 2011

disorganization) dimensions of any bargaining system (Traxler, 2003).
Second, it stresses actors’ strategies and interactions around domestic
struggles (endogenous factors) instead of exogenous pressures; thus
EMU is treated simply as one contextual variable. Third, it also focuses
on exchange mechanisms underlying policy concertation to explain the
likelihood and direction of institutional change. This reverses the causal
account in mainstream industrial relations studies (institutions shape
actors’ strategies) (Ferner and Hyman, 1998), by stressing the role of
actors’ strategies — particularly those of trade unions — in driving insti-
tutional change. Most studies of changes in collective bargaining during
the 1990s have neglected the fact that institutional change responds
mainly to processes of confrontation among actors. These processes open
the door to strategic realignments and rent-seeking behaviour. Because of
the apparent structural weakness of trade unions, all the scenarios above
have downplayed their capacity to act strategically in the political arena
and use this resource to revitalize their action in the industrial arena
(Frege and Kelly, 2003).
Bargaining system reform in Italy and Spain has been characterized by
social dialogue and policy concertation, as well as by diverse and
changing patterns of interaction based on exchanges between corporate
actors (Pizzorno, 1977) and on state intervention. While in Italy the
process has involved strong state intervention and extensive tripartite
policy concertation (state-sponsored reform), in Spain unilateral state-
imposed reform was followed by bipartite processes of narrow, ad hoc
and highly voluntaristic policy concertation and a reform negotiated
autonomously by unions and employers. This article stresses the
importance of unions’ strategic orientations regarding forms of partici-
pation in public policy-making; they have profited from the oppor-
tunities offered by the political framework to negotiate a new
compromise on collective bargaining structures in line with their
strategic orientations (Hamann and Martínez Lucio, 2003; Locke and
Baccaro, 1996). This change has not led to decentralization, but neither
has it reflected EMU-led consensus, or the absence of conflict and the
convergence underlying in the varieties of capitalism and coordination
literature. Instead, the analysis will show the existence of conflicting
views among actors regarding the direction of reforms. Observed differ-
ences between these two countries do not simply perpetuate pre-existing
institutional asymmetries, but reflect the outcomes of domestic
processes of negotiation and interaction. A trend towards greater
coordination of wage bargaining at national level can be observed in
both cases, but there persist differences as to the underlying mechanisms
as well as to the vertical dimension of articulation.
This article begins with a brief discussion of the political economy of
wage bargaining reform, with a focus on the sources of change and the
Molina Romo: Political Exchange and Bargaining Reform
9
02 romo (ds) 21/1/05 10:27 am Page 9
peer-00570929, version 1 - 1 Mar 2011

way in which exogenous pressures interact with endogenous factors to
deliver reform. It then analyses the experience of reform in the two coun-
tries between 1991 and 2001, with a focus on the politics of institutional
reform, the role of political exchange and the outcomes of the reforms.
The Political Economy of Wage Bargaining Reform:
Concertation and Political Exchange in Italy and Spain
Analyses of institutional change in labour markets during the last decade
emphasized the role of internationalization and the EMU. The latter was
expected to cause changes in bargaining structure in order to link wage
increases more closely to productivity growth. The 1990s indeed
witnessed transformations in bargaining structures: in the predominant
locus of bargaining (with a shift to the company level), the relationship
between different levels and the substantive scope of collective bargain-
ing (European Commission, 2001). The literature has produced a series
of analytical polarizations to map these transformations: Europeaniza-
tion or renationalization (Martin, 1999), centralization or decentraliza-
tion, coordination or disorganization (Soskice, 1990), organized or
disorganized decentralization (Traxler, 1996).
Such dichotomies are not applicable in the cases of Italy and Spain for
two main reasons. The first concerns the dependent variable. Most
studies have focused only on the size and scope of the bargaining unit,
neglecting more complex aspects of bargaining structure. They have
failed to consider the concept of articulation (Crouch, 1992), which
denotes the vertical and horizontal connections between units and actors
in the system. The articulation of bargaining systems depends in part on
structural factors (Rodríguez, 2000) such as the legal regulation of collec-
tive bargaining, the existence of extension mechanisms, economic
conditions and industrial structure. It also reflects the bargaining
strategies of the actors and their organizational capacities.
The second reason why existing analysis performs poorly relates to the
sources of change, i.e. the independent variable. Most authors anticipated
that exogenous factors (EMU) would cause convergence towards greater
deregulation and decentralization as a consequence of pressure for
moderation and competitiveness reinforced by employers’ increased
bargaining power. Other authors have predicted a convergence in policies
as well as in outcomes; for example Pérez (2002) diagnosed a parallel
process of reform in Italy and Spain. Both bargaining systems previously
displayed excessive informality and fragmentation resulting in high wage
increases and, in the context of restrictive monetary policy, provoked
rising unemployment. With the approach of EMU, the actors attempted
to reorganize the bargaining structure in order to attain greater internal
European Journal of Industrial Relations 11(1)
10
02 romo (ds) 21/1/05 10:27 am Page 10
peer-00570929, version 1 - 1 Mar 2011

coordination at the national sectoral level, while at the same time
rendering it more consistent with the policies of a new centralized
monetary authority, which would challenge institutions in a similar way.
These processes favoured consensus-led change.
Evidence, presented in this article, forces us to reconsider these views
by putting greater emphasis, first on unions’ collective bargaining
strategies and participation in policy-making, and second on the charac-
ter of interaction between actors. This is because the structure of collec-
tive bargaining is a contested institution, resulting from a political
confrontation between corporate social actors: their strategies and prefer-
ences are shaped by their relative power, which is reconfigured through
their interaction and the underlying exchanges in the political arena. The
reform of collective bargaining in the run-up to EMU was seen as a
common good, but its realization posed inter-associational problems
which could only be resolved through political exchange. Therefore, the
article adopts a ‘Pizzornian’ view of institutional change: notwithstand-
ing the existence of strong exogenous pressures imposing precise require-
ments for change, the resulting domestic articulation of bargaining is
endogenous to the system itself, being the result of an interaction based
on exchange between unions, employer organizations and governments.
Institutions, economic and political conditions at time t endow actors
with a set of incentives and resources to change the system: the articu-
lation of the structure at t+1 will reflect differences in power and
exchanges between the actors that participate in the struggle for insti-
tutional change.
By the end of the 1980s, both the Italian and Spanish collective
bargaining systems were subject to informal decentralizing pressures.
The absence of formal regulations concerning the articulation of
bargaining structures in Italy, and the lack of clear criteria in Spain, left
both systems dependent on changes in the bargaining power of actors.
After the incomes policies of the early 1980s, the Spanish bargaining
structure suffered from the absence of interconfederal agreements, the
increasing importance of the regional (provincial) level and the lack of
bargaining coverage for small companies. In Italy, employers tried to
bypass national sectoral agreements (contratti collettive nazionali di
lavoro or CCNL) in order to negotiate wages only at firm level. Trade
unions in both countries expressed their willingness to reform the
bargaining structure. The largest confederations in Spain, CC.OO. and
UGT, wanted to consolidate the national sectoral level of bargaining,
rather than the provincial level, in order to avoid further fragmentation.
In Italy the main unions, CGIL, CISL and UIL, were more concerned
with the development of bargaining structures at the local level, but
under the umbrella of CCNL. In both countries, employers (CEOE in
Spain, Confindustria in Italy) demanded a simplification of bargaining
Molina Romo: Political Exchange and Bargaining Reform
11
02 romo (ds) 21/1/05 10:27 am Page 11
peer-00570929, version 1 - 1 Mar 2011

levels, regulation and procedure. Accordingly, by the beginning of the
1990s the bargaining systems in Italy and Spain had generated different
incentives for each of the actors to introduce changes. In 1989 the
Spanish unions adopted a joint strategy, rejecting grand social pacts and
focusing on the qualitative and quantitative extension of collective
bargaining. Italian unions also developed unity of action and a focus on
the consolidation of their role as political actors (Baccaro et al., 2003);
however, internal divisions, and the distinctive political and institutional
context, entailed that negotiated adjustment differed both substantively
and procedurally from that in Spain.
Italy: State-sponsored Reform
At the end of the 1980s, the Italian bargaining system exhibited several
shortcomings. It was very informal, making it highly unpredictable and
voluntaristic (Regalia and Regini, 1998). Second, there were several
instances of bargaining being undertaken without a clear distribution of
tasks. The 1983 social pact (the ‘Scotti protocol’) confirmed the principle
of separate negotiations at three levels (Mariucci, 1985) without any clear
hierarchy. This allowed an employer-led process of disorganized decen-
tralization during the 1980s: the extension of company-level micro-
corporatism, with the strategic paralysis of national unions enabling a
gradual shift from national sectoral agreements towards local union nego-
tiations. From 1988 onwards, the public sector assumed wage leadership,
with disruptive consequences for economic performance. The contents
of collective agreements were very limited, and the system generated an
excessively compressed wage structure.
In this context, every actor had incentives to reform the structure of
collective bargaining. Unions conceived reform as a means of avoiding
further fragmentation within the trade union movement by making the
system more responsive to an increasingly heterogeneous labour force,
alleviating the decreasing representative capacity of confederal
unionism by establishing an articulated structure that could link
company bargaining with national federations, and strengthening
unions’ company-level presence by widening the content of negotia-
tions. Confindustria considered the wage determination system to be
too rigid and complicated to meet the differentiated wage requirements
of Italian firms and the variability in local labour markets. They also
criticized the informality and lack of articulation between levels of
collective bargaining, which diminished the predictability of its
outcomes. Finally, employers wanted changes that could gradually
reduce inflation and labour costs.
European Journal of Industrial Relations 11(1)
12
02 romo (ds) 21/1/05 10:27 am Page 12
peer-00570929, version 1 - 1 Mar 2011

The 1993 Pact: Competitive Decentralization of Bargaining and
the Importance of Emergency Political Exchange
One year after Confindustria unilaterally repudiated the long-established
wage indexation system (scala mobile), the 1992 incomes policy agree-
ment formalized its abolition, imposed a moratorium on company-level
bargaining and initiated negotiations to reform collective bargaining.
However, the position of the parties diverged substantially: the unions
sought a process of coordinated decentralization, with four-year sectoral
agreements and (in the case of CGIL and CISL) an obligation to bargain
at company level; Confindustria sought a disorganized process, down-
grading the role of the CCNL and exempting smaller firms from the obli-
gation to bargain. Several months of meetings failed to bring agreement,
and because of the critical economic juncture the Ciampi government
intervened. Its proposals for a pact to sustain economic recovery were
accepted: the unions obtained the formalization and consolidation of a
two-tier bargaining system, with guidelines set at interconfederal level,
while the employers were given financial incentives to reduce non-wage
labour costs and obtained greater employment flexibility.
The July 1993 Social Pact affected the articulation of collective bargain-
ing in two main ways: it created new workplace representation structures,
the RSU (rappresentanze sindacali unitarie); and it established new rules
governing relationships between bargaining levels. It thus reduced the
previous fragmentation and established an articulated structure, defining
rules governing the way the actors interact at the several levels. However,
it did not deliver a wholly articulated system of collective bargaining; it
failed to end disorganized renegotiation at company level of wages and
conditions agreed in the CCNL.
The collective bargaining system agreed in July 1993 was thus the
result, not so much of truly consensual tripartite negotiations as of a
government ‘exchange’ that the collective actors endorsed. This helps
explain why the system was so heavily contested afterwards: it did not
reflect a balanced equilibrium between the preferences of unions and
employers. Thanks to their unity of action, unions made the most of the
rather exceptional economic and political conditions of those years to
put pressure on the government, hence obtaining a document that suited
their preferences and interests. Employers’ fears of social unrest and the
breakdown of consensus in a period of economic crisis and political
instability increased the bargaining power of the unions and the regu-
latory power of concertation. The government initiated a successful
exchange between short-term wage moderation and the acceptance
of economic restructuring in exchange for the introduction of long-
term reforms in the collective bargaining system in line with union
preferences, as well as a more formalized and stable participation of
Molina Romo: Political Exchange and Bargaining Reform
13
02 romo (ds) 21/1/05 10:27 am Page 13
peer-00570929, version 1 - 1 Mar 2011

trade unions and employers’ organizations in national macroeconomic
management (Alacevich, 2000).
Formalization and Controlled Decentralization of Collective
Bargaining
Negotiations in 1998 failed to tighten the structure. Confindustria, and
the government, sought to reduce the scope for two-tier bargaining,
while CGIL and UIL insisted on retaining the broad contours of the
1993 agreement. This showed, in retrospect, that only exceptional
political and economic conditions in 1992–93 had enabled agreement on
bargaining reform. The ‘Christmas Pact’ finally agreed in December
1998 reaffirmed the actors’ commitment to concertation over macro-
economic and labour market policies, but there was no consensus on
bargaining structure. The only change was that the wage guidelines
approved at interconfederal level would follow the average inflation
rate of the EU, rather than forecast inflation in Italy. The 1993 and 1998
pacts have established different mechanisms at national, regional and
local levels through which interaction takes place; negotiations on a
fully integrated bargaining structure failed because of differences
between the trade unions, and the passing of the political and economic
emergency.
Confindustria reopened the debate in 1999, seeking to reduce the scope
for two-tier bargaining as well as to introduce greater flexibility, leading
to a gradual decentralization and individualization of industrial relations.
The emphasis was now on reducing the role of the CCNL, and eventu-
ally eliminating it altogether. In particular, the CCNL should no longer
regulate wages, which should be solely determined at company level.
Even among the unions, consensus on defending the two-level bargain-
ing structure has broken down. According to CISL, the primary bargain-
ing role should have been shifted towards company and regional levels,
with the CCNL only defining minimum standards. CGIL, however,
defended the 1993 two-level system and demanded a reinforcement and
extension of the role of the CCNL.
Benefits for unions from the agreed changes have been numerous. By
confirming sectoral negotiations (where unions enjoy relative strength)
as the main locus of regulation, they could maximize their influence on
the outcomes of the collective bargaining process. The creation of new
and strengthened company-level structures with clear links to higher-
level structures have provided a remedy for the crisis of representative-
ness, and the dual and informal character of collective bargaining (Regalia
and Regini, 1998). Finally, according to the 1993 Pact, the regional-level
bargaining is substituted for company-level bargaining in those regions
where small companies (SMEs) predominate. This helps the unions to
European Journal of Industrial Relations 11(1)
14
02 romo (ds) 21/1/05 10:27 am Page 14
peer-00570929, version 1 - 1 Mar 2011

solve the problem of the lack of coverage of collective bargaining and
union presence in SMEs (Demekas, 1995).
These changes have formalized a relatively articulated two-tier pattern
of collective bargaining, with a new balance between flexibility and
solidarity. Accordingly, it is difficult to characterize the Italian experience
in terms of either centralization or decentralization: according to some
authors, these changes have simply continued tendencies apparent in
Italy during the 1980s (Bellardi, 1997). Overall, one can say that they have
initiated a process of controlled decentralization, with an increase in the
importance of company-level bargaining but within the regulatory
framework set by the CCNL (Bordogna, 1997). Similarly, these changes
have redefined two balances underlying the Italian bargaining system.
First, that between the ‘micro’ requirements for higher wage flexibility
and competitiveness and the ‘macro’ need for stability and predictability.
The former has been achieved through strengthening company-level
bargaining (Origo, 2000), while stability and predictability derive from
the guidelines for wage increases issued by national sectoral and inter-
sectoral negotiations. Second, there is a new equilibrium between flexi-
bility and solidarity in wage policies: while there has been a real extension
of performance- and productivity-related wage clauses at company level
(Rapporto CNEL, 2001), the CCNL ensure that increases are negotiated
within similar bands for all categories of employees in most firms.
Spain: From State-imposed to Interconfederal Reform
With the coming of democracy in Spain, there was a shift from a statist
to a voluntaristic industrial relations system with a constitutional recog-
nition of the role of unions to represent the economic and social inter-
ests of workers. The Workers’ Statute of 1980 confirmed this, and
reaffirmed a collective bargaining system based on the autonomy of the
collective actors.
During the 1980s, collective bargaining displayed little articulation and
excessive informal decentralization (Valdés, 2001); bargaining took place
at several levels, but the main locus was the province. Framework sectoral
agreements signed at national or provincial level set the basis for lower-
level negotiations. The main problem consisted in the limited scope of
collective bargaining: first because negotiations at regional- or national-
sectoral level only occasionally affected workers in SMEs, where
unionism was absent; and second, because the issues discussed were, with
few exceptions, limited to pay and working time (Miguélez and Rebollo,
1999). Coverage of collective agreements was, however, high compared
to the low levels of union density, as a result of mandatory extension
mechanisms. Overall, the Workers’ Statute established a system very
Molina Romo: Political Exchange and Bargaining Reform
15
02 romo (ds) 21/1/05 10:27 am Page 15
peer-00570929, version 1 - 1 Mar 2011

sensitive to changes in the strategies or preferences of actors, hence
lacking stability and a source of permanent conflict.
Wage determination was shaped by rigid legal regulation. A statutory
minimum wage was set every year by the government, and sectoral
collective agreements set a basic minimum over this figure, serving as a
floor for lower-level negotiations. Nonetheless, bargaining at company
level was largely restricted to the negotiation of fringe benefits, as sectoral
(provincial or national) agreements exhausted determination of most
components of pay. Until 1994, pay structure was determined by regu-
lations dating back to the 1970s. At the beginning of the 1990s, the system
still displayed most of the shortcomings of the Franco regime.
State-imposed Disorganized Decentralization and
Individualization in 1994
The first reform of collective bargaining after the 1980 Workers’ Statute
occurred in 1994. Following several failed attempts to negotiate an agree-
ment with unions and employers, in 1993 the Socialist government
presented to the Consejo Económico y Social (CES, Economic and Social
Council) a proposal for labour market reform. This sought to strengthen
the connections between the different levels of wage determination.
However the government was weak (its majority depended on precarious
support from regional-nationalist parties, and it was to lose power in
1996) while the unions opposed major changes. What was agreed in 1994
reform was a process of decentralization and ‘the extension of a differ-
entiating and asymmetric microcorporatism’ (Sanguineti, 1999). The
changes strengthened collective bargaining as the main regulatory source
for labour relations through the abolition of the old Labour Ordinances
(mandatory regulations of working conditions) and increased the
bargaining autonomy of the collective actors.
Trade unions reacted by trying to fill the vacuum in order to avoid
further deregulation and disorganization. Nonetheless, there was an
inherent contradiction: while the government explicitly expressed its
intention to open new spaces for collective bargaining, it regulated the
way in which this should be structured, entailing a ‘de facto restataliza-
tion’ of the bargaining system that reduced its external autonomy
(Villacencio, 1998: 98). By contrast, the reforms also reinforced the
disorganized decentralization and deregulation of working conditions
already initiated in the late 1980s (López, 2000). Overall, we can say that
this reform entailed decentralization without articulation, substituting
rigidity by disorder.
This reform consolidated most of the defining features of wage
determination today. In particular, it formalized the increase in variable
(productivity-linked) wages at company level as against higher-level
European Journal of Industrial Relations 11(1)
16
02 romo (ds) 21/1/05 10:27 am Page 16
peer-00570929, version 1 - 1 Mar 2011

regulation (Fina et al., 2001). There was thus a process of disorganized
decentralization: companies were virtually unconstrained in the terms
which they could negotiate locally. This was strongly criticized by the
unions, which saw it as an attack on their power and role in the collec-
tive bargaining system. It encouraged a disorganized collective bargain-
ing structure, which failed to match the existing structure of union
organization; it strengthened the role of autonomous regional unions;
and it empowered employers to modify unilaterally the conditions nego-
tiated at multi-employer level. Accordingly, in the weeks prior to the
approval of the changes in parliament, the unions called for a general
strike. However, the government mounted an effective media campaign,
insisting that the reforms were essential if Spain wanted to join EMU
(Chari, 2001: 66). Outflanked, the unions then decided to change their
strategy, now threatening to bargain aggressively — resulting in uncoor-
dinated conflicts — if all the contents of the reform were applied.
Even though the employers’ associations had supported the govern-
ment proposals, the threat of escalating wage conflicts led them to reach
a tacit agreement with the unions. They would not exploit the more
damaging elements of the reform, in particular those allowing derogation
from multi-employer agreements and increased flexibility of working
conditions; and they would cooperate to achieve greater order in the
collective bargaining structure. In exchange, the unions agreed to
continued wage moderation and a commitment to reduce conflict
(Bentolila and Jimeno, 2002: 10). In consequence, the reform did not
provoke across-the-board decentralization and deregulation (Del Rey,
1996: 109–10). Thus the potentially decentralizing wage opt-out clause
did not produce the expected results in terms of pegging wages to
productivity. Empirical evidence shows its marginal effects on the articu-
lation of collective bargaining and on the process of wage determination
(Villacencio, 1998: 119). Overall, the 1994 reform failed to produce a
more articulated collective bargaining structure.
Policy Concertation and Political Exchange in 1997: Coordinated
Re-centralization
The new right-wing government of the Partido Popular (PP) initiated
in 1997 a process of tripartite social dialogue on reform of the labour
market. This resulted in three interconfederal agreements (acuerdos
interconfederales): para la estabilidad en el empleo (on employment
stability, AIEE), sobre negociación colectiva (on collective bargaining,
AINC) and sobre cobertura de vacíos (on filling the gaps in collective
bargaining, AICV). The AINC helped speed up and give greater
coherence to the replacement of the fascist Labour Ordinances by
collective agreements, and also dealt with bargaining articulation; the
Molina Romo: Political Exchange and Bargaining Reform
17
02 romo (ds) 21/1/05 10:27 am Page 17
peer-00570929, version 1 - 1 Mar 2011

AICV focused exclusively on extension of the coverage of collective
bargaining.
The approach adopted by the new government was crucial for success
in the negotiations. First, because it demonstrated considerable appre-
hension about the reaction of the unions to its economic policy (Soto,
2000: 71). Second, because the moderate stance adopted by the new
labour minister, who initiated negotiations on the basis of a prior under-
standing between unions and employers, created space for autonomous
bipartite negotiations. Finally, by initiating negotiations on a range of
labour market issues the government opened the possibility of exchange,
as indicated below.
The preferences of unions and employers regarding collective bargain-
ing reform were polarized. The unions wanted to improve articulation,
so as to rationalize the distribution of tasks between the different levels
and loci of negotiation and hence avoid permanent renegotiation as well
as informal decentralization. Any reorganization had to centre on
national-sectoral negotiations. The employers, by contrast, did not want
to modify the collective bargaining system, even though it was very
conflict-prone. The flexibility already introduced by the 1994 reform
provided them with a strong instrument to perpetuate the structural
weakness of the unions, allowing employers considerable freedom to
determine working conditions unilaterally.
The exchange underlying the simultaneous negotiation of the AIEE,
AINC and AICV enabled an agreement. The trade unions accepted less
stringent legal constraints on dismissals, in return for a reform of the
collective bargaining system more in line with their own preferences,
together with a limitation on the use of fixed-term and temporary
contracts and stronger incentives for the use of permanent contracts. The
agreement was important for the government not only to gain legitimacy,
but also to prevent the centrifugal tendencies that had occurred as a
consequence of the 1994 labour market reform: it was in the interest all
three social partners to bypass the regional veto (AARRII, 1996: 1228).
Unlike the 1994 reform, that of 1997 rationalized and improved the
articulation of the collective bargaining system. The starting point for
negotiations was similar to that three years earlier: the system remained
extremely complex since it involved negotiations at various levels and still
lacked a clear definition of the responsibilities at each of these, often
resulting in the renegotiation of the same issue and in inconsistencies. The
1997 agreements made two main contributions to reform. First of all, the
AINC enabled a real and effective increase in the regulatory space for
collective bargaining. It increased the power and autonomy of the
sectoral federations in collective bargaining, as against the national
confederations and regional organizations. The accord defines the
national sector as the ‘ordinary unit of bargaining’, and allows for sectoral
European Journal of Industrial Relations 11(1)
18
02 romo (ds) 21/1/05 10:27 am Page 18
peer-00570929, version 1 - 1 Mar 2011

differentiation of bargaining structures. Second, it has led to an improve-
ment in the articulation of the system through the development of a more
centralized and hierarchically ordered bargaining model. The objective is
to decrease labour market fragmentation (by guaranteeing some degree
of uniform regulation) while at the same time maintaining and improv-
ing company competitiveness. Nevertheless, the industrial relations
actors have encountered several difficulties in implementing this reform:
weakly structured employers’ organizations, problems in redistributing
bargaining power within the unions’ structures, and inconsistencies
between the terms of the new reform and what was legally established in
1994 and still remained in force (Baylos, 2002: 212).
This reform had important implications for wage determination. First,
the unions accepted an extension of the use of variable pay in exchange
for clear regulation by national sectoral agreements. Second, bargaining
was reorganized around these agreements, which now defined the
payment structure and set a minimum wage threshold for lower level
negotiations (Goerlich, 1997: 112). The AINC established that sectoral
agreements should be restricted to these issues, while increases for each
component of the wage structure should be set at lower levels.
The Failure of the 2001 Attempt at Revision
The 1997 agreement specified that an assessment and revision of the
contents of both the AIEE and AINC would be made in 2001. Accord-
ingly, further reform of the labour market and the collective bargaining
system was included in the agenda set out by the PP government after its
re-election in 2000. The parties failed to reach agreement on the reform
of the labour market, and the government acted unilaterally; but the
unions and employers continued negotiations over the reform of collec-
tive bargaining. In contrast to the process in 1997, while the government
gave an initial impulse to the negotiations it also threatened to intervene
if collective actors did not reach an agreement.
The unions and employers had sharply opposing aims when negotia-
tions started in June 2001. The unions wanted to remedy some of the
adverse consequences of the 1994 reform through a process of organized
decentralization. Both confederations agreed on the need to resolve the
difficulties they experienced in implementing the AINC, in consequence
of their organizational weakness at company level. Accordingly, their
priority was on finding mechanisms for an organized extension of collec-
tive bargaining to new sectors as well as to SMEs. The national sectoral
level would maintain a key regulatory role, while regional agreements
would cover those companies lacking an agreement (CC.OO., 2001;
UGT, 2001). The employer organizations also pushed for further decen-
tralization, but unlike the trade unions they demanded a reduced role for
Molina Romo: Political Exchange and Bargaining Reform
19
02 romo (ds) 21/1/05 10:27 am Page 19
peer-00570929, version 1 - 1 Mar 2011

both sectoral and regional levels. They also rejected the mandatory exten-
sion of agreements. Finally, the government expressed its preference for
a law defining the issues to be negotiated at each level, with a minor role
for provincial agreements and an extension of opt-out clauses (El País, 20
February and 25 June 2001).
This antagonism precluded any progress in negotiations, and the
government threatened to intervene unilaterally. The unions and employ-
ers continued bipartite negotiations, eventually signing a new intercon-
federal accord (AINC-2002). This was the first interconfederal agreement
on wage moderation to be signed since 1984. However it failed to intro-
duce reforms in the structure of collective bargaining, merely establish-
ing joint criteria for moderation in wage bargaining in 2002 in the
interests of employment creation and company competitiveness. The
agreement was possible only on the basis of an exchange between a
commitment to greater stability of employment (including clauses to
neutralize through collective negotiations some of the harmful impli-
cations of the labour market reform imposed by the government) and the
promotion of negotiated flexibility with moderate wage increases and a
compromise to keep conflict at low levels. Thanks to this agreement, the
unions prevented the unilateral imposition of the government proposals
and maintained the wage bargaining structure established in 1997
(Alarcón, 2001).
Conclusion: Bargaining Articulation through Political
Exchange
This article has shown how formal changes in the structure of collective
bargaining and wage determination during the last decade in Italy and
Spain have been the outcome of the exchange underlying patterns of
consensus and conflict between unions, employers’ associations and
governments. Despite most predictions of the impact of EMU on indus-
trial relations, collective bargaining structures in Italy and Spain have not
moved towards a model of market-driven decentralization and individu-
alization. Political exchange has played a key role in determining the
direction of reforms, by providing a mechanism for coordinating actors’
responses and channelling the contradictory forces challenging collective
bargaining systems. In particular, through policy concertation and an
underlying process of exchange, trade unions have managed to formalize
the collective bargaining system in order to halt processes of informal
decentralization, and to consolidate union-friendly arrangements. These
developments are summarized in Table 1.
Economic adjustment in the 1990s confronted confederal unions in the
two countries with more threats than opportunities. The problems were
European Journal of Industrial Relations 11(1)
20
02 romo (ds) 21/1/05 10:27 am Page 20
peer-00570929, version 1 - 1 Mar 2011

accentuated by their politicization, their declining representative capacity
and their fragmentation within weakly formalized and structured collec-
tive bargaining systems. Informality led to a process of de facto decen-
tralization during the 1980s, which proved very damaging to union
power and interests. The unions responded by participating in political
processes to promote collective bargaining reforms aimed at setting
clearer, formalized and articulated rules. Critical to render this strategy
effective and obtain the expected results has been unity of action among
confederal unions.
Thus in contrast to the events of the early 1980s, policy concertation
in both countries in the 1990s turned out to be a resource in the hands of
unions. During 1980–86 in Spain and 1983–84 in Italy, tripartite social
pacts had involved a short-term exchange that further weakened the
Molina Romo: Political Exchange and Bargaining Reform
21
TABLE 1. Political Exchange and Bargaining Reform
Character of reform Exchange
Spain 1994 Unilateral government Implicit ex-post bipartite exchange:
imposition unions accept wage moderation in
exchange for the non-application of
some of the clauses of the reform.
Spain 1997 Tripartite agreement Unions accept weaker dismissal
on labour market protection and (implicitly) wage
reform moderation in exchange for a
bargaining structure based on sectoral
agreements, as well as measures to
reduce temporary employment.
Spain 2001 Failed negotiations Implicit ex-post bipartite exchange:
on structural reform unions accept wage moderation in
but interconfederal exchange for the non-application of
agreement on some of the clauses of the government-
collective bargaining imposed reform.
Italy 1993 Tripartite social pact Unions accept wage moderation,
restrictive economic policy and
abolition of the scala mobile in
exchange for a formalization of dual-
level collective bargaining, the
consolidation of new company-level
representation structures and the
institutionalization of policy
concertation.
Italy 1998 Failed negotiations Differences between unions, together
on bargaining reform with lack of public resources for
but tripartite exchange prevent agreement on reform
‘Christmas pact’ of bargaining structure.
02 romo (ds) 21/1/05 10:27 am Page 21
peer-00570929, version 1 - 1 Mar 2011

position of the confederal unions. Wage moderation, the centrepiece of
these exchanges, was traded against compensation that failed to resolve
the challenges they faced as representatives of the labour force. Instead,
unions accepted short-term reinforcement of their role as political rather
than industrial actors, and paid more attention to the confederal level
than to grassroots structures. This, together with weakly formalized
collective bargaining systems that were undergoing informal processes of
decentralization, placed the unions in a critical situation at the beginning
of the 1990s.
Policy concertation during the 1990s was thus different, both in its
process and in the underlying exchange, especially in Spain. Here, trade
unions tried to escape from grand tripartite social pacts where the costs
are immediate and easily perceptible while the benefits are diffuse and
effective only over a longer time span. Instead, they pursued, targeted and
specialized social dialogue focused upon the negotiation of institutional
reforms. Reform of the collective bargaining system was one of the core
issues within this strategy. Three main goals were pursued: an extension
of the regulatory capacity of collective bargaining, a formalization of the
rules connecting levels within the system and the consolidation of the
national sector as the predominant bargaining level. Their opposition to
income policies, which were not accompanied by a change in macro-
economic policy, as well as to any grand social pact covering a range of
issues, led them to abandon discussions in 1993, which obliged the
government to reform the bargaining structure unilaterally. A favourable
political context resulting from the change in government allowed the
unions in 1997 to reach a successful and favourable exchange, thanks to
the parallel negotiations on labour market reform. Finally, opposition
from both unions and employers to a new unilateral intervention by
government left the collective bargaining structure unchanged in 2001.
In Italy, the 1993 reform came about as a result of an interconfederal
tripartite social pact. The unions pursued the consolidation of a bargain-
ing system with strong formal links between higher and lower level units
of negotiation. They managed to profit from the rather peculiar economic
and political conditions when the pact was signed, to formalize and
consolidate a two-tier structure, which not only reformed company-level
mechanisms of union representation but also established a clear distri-
bution of tasks between the different levels and clear procedures for the
participation of collective social actors in policy-making. In exchange,
the unions accepted wage moderation, a restrictive economic policy and
the abolition of the scala mobile (already announced a year before by the
employers’ organizations).
As a consequence of these processes of policy concertation and
political exchange, both the Italian and the Spanish structures of bargain-
ing are now endowed with clearer and more rational rules, as is shown in
European Journal of Industrial Relations 11(1)
22
02 romo (ds) 21/1/05 10:27 am Page 22
peer-00570929, version 1 - 1 Mar 2011

Figure 1. In Spain, attempts at strengthening the articulation of the collec-
tive bargaining structure have tended to shift the predominant locus of
bargaining to the national sectoral level. The 1997 reform has allowed the
national sectoral federations to recover their bargaining power, but has at
the same time preserved the capacity of firms to adapt the conditions set
at higher levels to their particular circumstances. Today the collective
bargaining structure displays greater articulation, with sectoral negotia-
tions setting the path to be followed by negotiations at lower levels. All
the reforms have extended the regulatory power of collective bargaining.
The specific distribution of tasks across levels within each sector has been
left to the sectoral federations since the 1997 reform, which means that
there is no single form of articulation. In Italy, we have seen the formal-
ization of a two-tier pattern of collective bargaining, with sectoral and
company levels at the centre of the system. The distribution of tasks
across levels was legally prescribed as part of the 1993 social pact, which
has resulted in a narrower autonomous sphere of regulation by unions
and employers than in Spain.
REFERENCES
Alacevich, F. (2000) Emergenza Occupazione: Ruolo delle Parti Sociali,
Concertazione e Contrattazione Collettiva in Italia. Florence: Cusl.
Alarcón, M.R. (2001) ‘Reflexión Crítica sobre la Reforma Laboral de 2001’,
Revista de Derecho Social 15: 35–49.
Molina Romo: Political Exchange and Bargaining Reform
23
No formal articulation.
Informality and
discretion in decisions
about selection of
bargaining levels
National / Provincial
Spain late 1980s
The distribution of
tasks is regulated
autonomously by
unions andemployers
National-Sector/
Spain 1997
No formal
articulation. Informal
de-centralization to
company
National-Sector/
Company
Italy late 1980s
Formal regulation of
task distribution
across levels.
Articulated/
controlledde-
centralization
National-Sector/
Company
Italy 1993
Interconfederal
agreements establish
wage increases, but
the system remains
disorganized
Interconfederal/
Provincial
Spain1979–86
Interconfederal
agreements establish
wageincreases, but
the system remains
disorganized
Interconfederal/
National-Sector
Italy Early1980s
The distributionof
tasks across levels is
legally regulated
Bottom-up and top-
bottom
Provincial/ Company
Spain 1994
Horizontal
Disorganisation Coordination
Vertical
Centralization Decentralization
.
-
.
.
--
..
.
Horizontal
Disorganisation Coordination
interactions
Company
FIGURE 1. Vertical and Horizontal Dimensions of Articulation
02 romo (ds) 21/1/05 10:27 am Page 23
peer-00570929, version 1 - 1 Mar 2011

AARRII (1996) ‘Resaca Electoral: Los Pactos Posibles’, Relaciones Laborales 1:
1283–96.
Baccaro, L., Carrieri, M. and Damiano, C. (2003) ‘The Resurgence of the Italian
Confederal Unions: Will it Last?’, European Journal of Industrial Relations
9(1): 43–59.
Baylos, A. (2002) ‘Diálogo Social y Negociación Colectiva: Una Narración’,
Revista de Derecho Social 17: 203–25.
Bellardi, L. (1997) ‘Il Difficile Percorso Verso nuovi Assetti Contrattuali’, in
L. Bellardi and L. Bordogna (eds) Relazioni Industriali e Contrattazione
Aziendale, pp. 13–56. Rome: FrancoAngeli.
Bentolila, S. and Jimeno, J.F. (2002) La Reforma de la Negociación Colectiva en
España. Working Paper FEDEA no.3.
Bordogna, L. (1997) ‘Un Decennio di Contrattazione Aziendale nell’Industria’,
in L. Bellardi and L. Bordogna (eds) Relazioni Industriali e Contrattazione
Aziendale, pp. 61–108. Rome: FrancoAngeli.
Chari, R.S. (2001) ‘The EU Dimensions in Economic Policy-Making at the
Domestic Level: Evidence from Labour Market Reform in Spain’, South
European Society and Politics 6(1): 51–74.
CC.OO. (2001) Cuadernos de Información Sindical 16: 7–8.
Crouch, C. (1992) ‘The Fate of Articulated Industrial Relations Systems: A
Stock-Taking after the “Neo-Liberal” Decade’, in M. Regini (ed.) The Future
of Labour Movements, pp. 169–87. London: SAGE.
Del Rey, S. (1996) ‘Una Década de Transformación del Sistema de Negociación
Colectiva y la Refundación de la Teoría Jurídica de los Convenios
Colectivos’, Relaciones Laborales 1: 178–210.
Demekas, D. (1995) ‘Labour Market Institutions and Flexibility in Italy’,
Labour 9(1): 3–43.
EIRO (1999) ‘The “Europeanisation” of Collective Bargaining’,
http://www.eiro.eurofound.ie/1999/07/study/index.html
European Commission (2001) Industrial Relations in Europe. Report 2001.
Luxembourg: OOPEC.
Ferner, A. and Hyman, R., eds (1998) Changing Industrial Relations in Europe.
Oxford: Blackwell.
Fina, L. González de Lena, F. and Pérez, J.L. (2001) Negociación Colectiva y
Salarios en España. Madrid: CES.
Frege, C. and Kelly, J. (2003) ‘Union Revitalization Strategies in Comparative
Perspective’, European Journal of Industrial Relations 9(1): 7–24.
Goerlich, J.M. (1997) La Reforma Laboral de 1997. Valencia: Tirant Lo Blanch.
Hall, R. and Soskice, D., eds (2001) Varieties of Capitalism: The Institutional
Foundations of Comparative Advantage. Oxford: Oxford University
Press.
Hamann, K. and Martínez Lucio, M. (2003) ‘Strategies of Union Revitalization
in Spain: Negotiating Change and Fragmentation’, European Journal of
Industrial Relations 9(1): 61–78.
Kauppinen, T., ed. (1998) The Impact of EMU on Industrial Relations in
Europe. Helsinki: Finnish Industrial Relations Association.
Kerr, C., Dunlop, J.T., Harbison, F.H. and Myers, C.A. (1960) Industrialism
and Industrial Man. Cambridge: Harvard University Press.
European Journal of Industrial Relations 11(1)
24
02 romo (ds) 21/1/05 10:27 am Page 24
peer-00570929, version 1 - 1 Mar 2011

Kochan, T.A., Katz, H.C. and McKersie, C.B. (1986) The Transformation of
American Industrial Relations. New York: Basic Books.
Locke, R. and Baccaro, L. (1996) ‘Learning from Past Mistakes? Recent
Reforms in Italian Industrial Relations’, Industrial Relations Journal 27(4):
289–303.
López, E. (2000) La Descentralización de la Estructura de la Negociación
Colectiva y las Posibilidades de su Articulación. Valencia: Tirant lo
Blanch.
Mariucci, L. (1985) La Contrattazione Collettiva. Bologna: Il Mulino.
Martin, A. (1999) Wage Bargaining Under EMU: Europeanisation,
Re-Nationalisation or Americanisation. Working Paper, Harvard University
Center for European Studies.
Miguélez, F. and Rebollo, O. (1999) ‘Negociación Colectiva en los Noventa’, in
F. Miguélez and C. Prieto (eds) Las Relaciones de Empleo en España,
pp. 325–46. Madrid: Siglo XXI.
Molina, O. (2003) The Politics of Bargaining Reform in Italy and Spain
1991–2001. Working Paper, European University Institute, SPS 14.
Origo, F. (2000) ‘Chi Contratta il Premio di Risultato? Un’Analisi Empirica
Degli Effetti dell’Accordo di Luglio sulla Contrattazione Decentrata’, Lavoro
e Relazioni Industriali 1: 37–66.
Pérez, S. (2002) ‘Monetary Union and Bargaining Institutions in the EU.
Extrapolating from Some Member State Experiences’, Comparative Political
Studies 35(10): 1198–227.
Pochet, P., ed. (1999) Monetary Union and Collective Bargaining in Europe.
Brussels: PIE-Peter Lang.
Pizzorno, A. (1977) ‘Scambio Politico e Identità Colletiva nel Conflitto di
Classe’, Rivista Italiana di Scienza Politica 7(2): 165–98.
Regalia, I. and Regini, M. (1998) ‘Italy: The Dual Character of Industrial
Relations’, in A. Ferner and R. Hyman (eds) Changing Industrial Relations in
Europe, pp. 459–503, Oxford: Blackwell.
Regini, M. (2000) Modelli di Capitalismo. Le Risposte Europee alla Sfida della
Globalizzazione. Rome: Laterza.
Rodríguez Fernández, M. (2000) La Estructura de la Negociación Colectiva.
Valladolid: Lex Nova.
Sanguineti, W. (1999) ‘Libertad Sindical y Rol Institucional de los Sindicatos:
Anotaciones a dos Décadas de Vigencia del Texto Constitucional’, Revista de
Derecho Social 8: 35–59.
Scharpf, F. (1997) Games Real Actors Play: Actor-Centered Institutionalism in
Policy Research. Boulder: Westview.
Soskice, D. (1990) ‘Reinterpreting Corporatism and Explaining
Unemployment: Co-ordinated and Non Co-ordinated Market Economies’,
in R. Brunetta and C. Dell’Aringa (eds) Labour Relations and Economic
Performance, pp. 170–211, Houndsmills: Macmillan.
Soto, A. (2000) ‘Política social. Relaciones con los Sindicatos’, in J. Tusell (ed.)
El Gobierno de Aznar. Balance de una Gestión 1996–2000, pp. 71–96.
Barcelona: Critica.
Traxler, F. (1996) ‘Collective Bargaining and Industrial Change: A Case of
Disorganisation?’, European Sociological Review 12(3): 271–87.
Molina Romo: Political Exchange and Bargaining Reform
25
02 romo (ds) 21/1/05 10:27 am Page 25
peer-00570929, version 1 - 1 Mar 2011

Traxler, F. (2003) ‘Bargaining, State Regulation and the Trajectories of Industrial
Relations’, European Journal of Industrial Relations 9(2): 141–61.
UGT (2001) Reforma Laboral y Actuación de UGT en la Negociación
Colectiva. Criterios y Orientaciones. Madrid: Secretaria Confederal UGT.
Valdés, F. (2001) ‘Notas para una Reforma del Marco Legal de la Estructura
Contractual’, Revista de Derecho Social 15: 73–86.
Villacencio, A. (1998) Los Acuerdos Marco Sobre la Estructura de la Negociación
Colectiva: Naturaleza Jurídica, Eficacia, Contenido. Granada: Comares.
Wailes, N., Ramia, G. and Lansbury, R.D. (2003) ‘Interests, Institutions and
Industrial Relations’, British Journal of Industrial Relations 41(4): 617–37.
Weber, A.R. (1961) The Structure of Collective Bargaining. Chicago: Free Press
of Glencoe.
ÓSCAR MOLINA ROMO is a researcher at the European University Institute.
ADDRESS: Department of Social and Political Sciences, European University
Institute, Badia Fiesolana, Via dei Roccettini 9, 50016 San Domenico di
Fiesole (FI), Italy. [e-mail: oscar.molina@iue.it]
European Journal of Industrial Relations 11(1)
26
02 romo (ds) 21/1/05 10:27 am Page 26
peer-00570929, version 1 - 1 Mar 2011