The effects of the July 1993 tripartite agreement on company-level bargaining in 1995-6

Download article in original language : IT9805227FIT.DOC

April 1998 saw the release of the first figures for company-level collective bargaining in 1995-6, compiled by the Observatory on Company-level Bargaining in Italy (Osco). The results are of particular interest, as they refer to the period immediately following the reform of the Italian bargaining system introduced by the tripartite agreement of July 1993. The figures indicate that the 1993 agreement has had a positive effect in promoting company-level bargaining.

Since 1984, the Observatory on Company-level Bargaining in Italy (Osservatorio sulla contrattazione aziendale in Italia, Osco) has carried out periodic and systematic surveys of company-level bargaining in the main industrial sectors and certain service sectors (banks, commerce and transport). Starting from 1991, the survey is based on a nationally representative sample of companies with at least 20 employees. In April 1998, the Cesos (Centro di Studi Economici Sociali e Sindacali) research institute issued Osco's first figures for the two-year period 1995-6, which are of particular interest as this was the period immediately following the reform of the collective bargaining system introduced by the tripartite agreement of July 1993.

Company-level bargaining after the July 1993 tripartite agreement

The tripartite agreement of July 1993 made major changes to company-level (or "second-level") bargaining, providing it with a more definite framework than before, although without making it compulsory (for further details on the agreement's provisions, see IT9709212F and IT9803223F). The Osco survey for 1995-6 - which, in industry, covers around 650 companies and more than 200,000 employees - allows an examination of the effects of the 1993 reform on the dynamics and characteristics of company-level bargaining compared with previous trends, and especially with the experience of the early 1990s.

Simplifying greatly, the reform could have been expected to produce three main changes.

1. A significant increase in the proportion of companies and workers involved in decentralised bargaining (referred to respectively as bargaining propensity and bargaining extension). Company-level bargaining, in fact, was very rare in the years immediately prior to the reform. This was due not only to the economic difficulties of firms and of the Italian economy in general during those years, but also to the 18-month bargaining moratorium introduced in effect by a previous tripartite agreement in July 1992, and then to the de factopostponement of decentralised bargaining until after the first industry-level agreements had been renewed on the basis of the new system defined by the July 1993 agreement.
2. Greater regularity in the distribution of bargaining over time, in that the July 1993 reform provides for only one company-level agreement during the four years in which an industry-level agreement is in force. This should therefore have reduced the bargaining rate (average number of agreements per companies involved in bargaining). Moreover, this expected greater regularity suggests that, more than in the past, the dynamics of company-level bargaining should be evaluated across the entire period of validity of the industry-wide agreement. In the present case, therefore, assessment should also include 1997, whereas 1994 is less important because it was still a year of transition from the old to the new system.
3. The third, more qualitative, expectation concerns the issues on the bargaining agenda and, with particular reference to pay, the likely intensification of bargaining over the so-call "participatory wage" (salario partecipativo).

First results

The first results seem to confirm these expectations, as table 1 below indicates.

* Until 1990 the sample was not nationally representative because it was constructed on the basis of about 15 provinces (with slight variations in some years) distributed around the country but selected from the most industrialised areas. Since 1991, a nationally representative sample has been used.

** See main text for definition.

Source: Osco.

Bargaining propensity (the percentage of companies involved in bargaining on at least one occasion) grew in both years compared with 1994: to a slight extent in 1995 and more markedly in 1996, when 22% of the companies in the sample were involved in decentralised bargaining - a figure which is even higher than the annual average in 1991-3. When the two years are added together, the percentage amounts to around 35% of the sample, to which should be added the figures for 1997, for the reason mentioned above.

This increased propensity to bargain is associated with a larger percentage of workers covered by company-level bargaining (bargaining extension): slightly less that 20% in 1994, slightly more than 25% in 1995, and about 30% in 1996. This latter percentage is once again larger that the two immediately previous years, and similar to the annual average in 1991-3, but it is substantially lower than the percentage in the period 1984-90 (in which, however, the sample was different, being provincially based, so that the phenomenon was probably overestimated). If the two years are added together, the range of workforce coverage is around 60%. Finally, the bargaining rate (number of agreements per companies involved in bargaining) was still rather high in 1995, but fell in 1996.

However, these general values conceal a number of significant differences. Table 2 below breaks down the bargaining propensity and rate figures by industry, company size and the involvement or otherwise of trade unions.

*See main text for definition.

Source: Osco.

Leaving the sectoral figures aside, which also depend on the different times at which industry-level agreements have been concluded and hence company-level bargaining has begun, the most marked differences are to do with company size. This is an extremely strong relation between company size and bargaining propensity and rate, and it is invariably the most statistically significant factor in all the years surveyed by Osco since 1984. The percentage of large companies (more than 500 employees) involved in decentralised bargaining is between two and three (and in the past even four) times higher than that of small firms (between 20 and 99 employees). Workers employed in the latter are much more likely to be excluded from the advantages of company-level bargaining compared with their counterparts in companies with more than 500 employees. The same applies to the "negotiation rate": large companies bargain on more issues and more often.

The unionisation rate positively influences the likelihood of company-level bargaining, although in previous years the relation was non-linear - ie the presence of unions in the company had a certain effect, but above a certain threshold of unionisation the relation attenuated. Also to be noted is the close positive correlation of company-level bargaining with trade union involvement in company decision-making - through consultation/information, joint committees etc.

As regards bargaining issues, the most frequent of these are still, as in past years, trade union rights and related matters, working hours, the organisation of work, and pay. Bargaining on the last-named issue in particular intensified significantly in 1995 and 1996, as was partly to be expected (76% of companies bargained on wages in 1995, and 83% in 1996).

Commentary

The Osco survey indicates that, in the two years examined, company-level bargaining in industry covered approximately 35% of the companies in the sample and just under 60% of employees. If the figures for 1997 are added (although the bulk of bargaining probably took place in the two years surveyed), the coverage is even greater, and the positive effect of the July 1993 central tripartite agreement in promoting company-level bargaining is amply confirmed, in contrast to its pronounced decline during the early 1990s.

However, the figures also show a distinct difference between small and medium-to-large companies, with the employees of the former having two to three times less probability of being covered by company-level bargaining than the employees of the latter.

This finding highlights once again the profound dualism in Italian industrial relations, which any proposed revision of the bargaining system introduced by the 1993 agreement should bear in mind, especially with reference to the relationship between the national and decentralised levels of bargaining (Lorenzo Bordogna, University of Brescia)