Research examines employee representation

Download article in original language : FR0201111FFR.DOC

Late 2001 saw the publication by the Ministry of Employment of two studies of employee representation in French firms, based on widely-differing statistical sources. The Ministry published both the results of the 1999 works council elections and the findings of a survey of the existence of various employee representative structures in companies and workplaces. The two studies confirm that works councils are now a significant presence in the industrial relations landscape, and indicate a recovery in support for trade unions, particularly CFDT and CGT.

For several years, the methods of collecting statistical data on the collective representation of French employees, published by the Ministry of Employment, have been changing. New information derived from various sample-based surveys - such as 'Réponse' and the Survey on Labour Activity and Employment Status (Enquête sur l'Activité et les Conditions d'Emploi de la Main d'Oeuvre, ACEMO) - has supplemented the old sources such as election results for works council s, and industrial tribunal s (Conseils de prud'hommes - FR9710171F).

In late 2001, the Ministry of Employment's Office for Research and Statistics (Direction de l'animation de la recherche, des études et des statistiques du ministère de l'Emploi, DARES) published an analysis of the 1999 works council election results (Premières synthèses, No. 49.1, December 2001) as well as findings from an ACEMO survey on employee representative institutions in 1999 (Premières informations, No. 48.1, November 2001). The almost simultaneous publication of these two studies has highlighted the differences between the available data sets in terms of their nature, production methods and findings.

1999 works council election results

The DARES study of works council election results covers all companies or workplaces with at least 50 employees, which have a statutory obligation to hold works council (comité d'entreprise) elections every two years. These election results are sent by the concerned companies to the Ministry of Employment's decentralised offices. The results are often used to assess the strength of support for trade unions, which present slates of candidates for election.

The field of study for the DARES research is not directly comparable from one year to the next for various reasons. First, changes in company structure (closures, mergers, change of location etc) have an impact on representative institutions that is hard to deal with statistically. Second, since 1991, elections at the SNCF national railway have taken place in even years only. At SNCF, whose electorate accounts for almost 10% of all works council votes nationally, the turnout is higher than elsewhere, and most importantly, only unions are represented. This schedule for works council elections therefore justify a method of analysing the aggregate voting scores over a period of two consecutive years, ie on an 'election cycle' basis. This allows union support and its development to be charted more precisely. However, the results from an even or odd year are still comparable with those from two years previously - thus the 1999 results are comparable with those for 1997.

The 1999 election results corroborated prior developments (FR0011102F). The election turnout was slightly down on 1997 (FR9905185N), by 0.5 percentage points to 65.3 %. This trend is mostly due to large companies, where the drop was in the order of two points.

The most significant fact to arise from the 1999 results is the continuing decline of non-union slates, which can run in the second round of voting for works council members if there is a lack of union candidates in the first round. In the 1980s, union support was steadily eroded by non-union slates. Since 1992, this trend has been reversed. Between 1997 and 1999, non-union slates recorded their largest fall in support, by 3.5 points, thus falling below 26% of the vote. The larger the workplace or company, the higher the support for unions. The non-union slates, more often present in smaller workplaces or companies, experienced a considerable reduction in their vote (by six to seven points), in firms with fewer than 100 employees, which benefited the union slates, especially those of the General Confederation of Labour (Confédération générale du travail, CGT).

The relative support for the various unions has changed little. Since 1991, the French Democratic Confederation of Labour (Confédération française démocratique du travail, CFDT) has overtaken the CGT in odd years (ie those where no elections took place in the SNCF), with CGT leading both in even years and overall in each two-year cycle. This was also the case for 1999. Between 1997 and 1999, CFDT increased its vote by 2.5 points, to reach 22.9%, while CGT gained 1.1 points to 21.5% of votes cast. The French Christian Workers' Confederation (Confédération française des travailleurs chrétiens, CFTC) increased its vote by 0.7 points to 5.8%. The General Confederation of Labour-Force ouvrière (Confédération générale du travail-Force ouvrière, CGT-FO) and the French Confederation of Professional and Managerial Staff-General Confederation of Professional and Managerial Staff (Confédération française de l'encadrement-Confédération générale des cadres, CFE-CGC) retained their share of the vote at 12.2% and 6.3% respectively. The other unions unaffiliated to any of the five confederations with national representative status together lost 0.3 points of the overall vote, leaving their total support at 5.6%.

Representative bodies in 1999

The DARES survey of employee representative bodies and structures, carried out in 1999, used a representative sample of 11,000 workplaces and 900 businesses with at least 10 employees covered by the ACEMO field of study. Central and local government and much of the health and social service sectors are excluded . A questionnaire, sent to companies, dealt with the presence of various representative structures in the workplace or company being surveyed. These were:

• workforce delegate s (délégués du personnel),

• works councils;

• workplace health and safety committee s (comités d'hygiène, de sécurité et des conditions de travail, CHSCTs); and

• trade union delegate s (délégués syndicaux).

The findings of the survey show the extent to which the workplaces and companies concerned are covered by one or other of these institutions. A strong hypothesis in the study is that if one workplace or company is covered by a works council, all the workplaces within that company are also covered. The same assumption holds for trade union delegates and CHSCTs. The survey covers the period March 1997 to March 1999, in order to fit in all the works councils and workforce delegates elected on a two-year mandate.

In 1999, more than half of workplaces with at least 10 employees, which account for one in five employees, were not covered by any representative structure. However, collective employee representation is becoming a general rule in companies and establishments above the statutory threshold for setting up works councils. Fewer than 7% of workplaces with 50 employees or more now have no representative body, while 98% of those with over 250 employees have at least one form of representation. Moreover, the presence of union delegates leads to that of other representative structures. Lastly, the survey bears out the respective positions of the various unions: CFDT and CGT have the equal highest support.

Commentary

The production of French industrial relations statistics is paradoxical, in that there are few sources compared with other fields (such as employment), while at the same time the new sources used to increase the amount of data make them more confusing. Serious gaps between sets of findings have been identified, depending on the various sources. Thus, the hypothesis formulated in the ACEMO survey of employee representation in terms of the rate of coverage by such bodies, apart from being legally and sociologically questionable, might be interpreted differently by the companies surveyed.

In order to overcome these problems, a series of seminars involving researchers and trade unionists, arranged on the initiative of DARES and the Economic and Social Research Institute (Institut de recherches économiques et sociales, IRES), is currently assessing the available data and their possible development. (Catherine Vincent, IRES)