1998 works council election results published
Published: 27 November 2000
The results of the works council elections held in France in 1998 were published in October 2000. The main trends observed were a fall in turn-out among employees and a drop in the vote for non-unionised independent candidates. Nevertheless, independents still constitute the largest single group of works council members. The election results highlighted a number of problems in collecting data and in using these figures as a means of measuring the representativeness of trade unions.
Download article in original language : FR0011102FFR.DOC
The results of the works council elections held in France in 1998 were published in October 2000. The main trends observed were a fall in turn-out among employees and a drop in the vote for non-unionised independent candidates. Nevertheless, independents still constitute the largest single group of works council members. The election results highlighted a number of problems in collecting data and in using these figures as a means of measuring the representativeness of trade unions.
Private sector workplaces with more than 50 employees hold elections every two years for employee representatives on works council s (comités d'entreprise), which have a range of information, consultation and other rights. These elections are spread over the entire year, rather than being concentrated in a short period, as is the case in some other countries. The results are then gathered by the Ministry of Employment and Solidarity. They are often used to assess the strength of union support in companies of this size. October 2000 saw the publication of the 1998 election results by the Ministry's Office of Research Coordination and Statistics (Direction de l'animation de la recherche et des études statistiques, DARES) ("Les élections aux comités d'entreprise en 1998", Nadine Ruelland, DARES, Premières Informations No. 41-1, October 2000). In order to draw comparisons, the results from even years must be examined together, as must those from odd years. The new data deal with even years, and should thus be compared with 1996 (FR9809131F), while last year the Ministry published the results for 1997, an odd year (FR9905185N). The full electoral cycle can be studied by adding two consecutive years' results (currently 1997 1998, which can then be compared with either 1995 1996 or 1996 1997).
Lower turnout and fall in vote for independent candidates
Comparison between the 1998 and 1996 results reveals two main developments. First, the turnout among employees dropped by 0.6 percentage points, to 65.7%. This decrease confirms a trend that has been in progress for almost 30 years. Yet this fall is cancelled out and even reversed when the results from temporary work agencies and companies which employ many staff on a short-term basis (as in retail, for example) are taken out of the figures. Excluding these firms, there is a 0.5 point rise in turnout, bringing it up to 70.1% among the remaining companies. A proportion of the fall in turnout may thus well be due to the poor voting record of workers in precarious employment.
Second, the vote for independent candidates (ie those not affiliated to trade unions) fell in 1998, thus continuing a trend which began seven years ago. Independent slates won 24.1% of the vote in 1998, compared with 25.1% in 1996 - see table 1 below. This trend is all the more pronounced because changes in the distribution of votes between trade union candidates were modest.
| . | 1996 | 1998 | 1995-6 | 1997-8 |
| Turn-out | 66.3 | 65.7 | . | . |
| CGT | 23.6 | 24.3 | 21.7 | 22.5 |
| CFDT | 21.5 | 21.7 | 21.1 | 21.2 |
| CFTC | 4.5 | 4.9 | 4.8 | 5.0 |
| CGT-FO | 12.1 | 12.1 | 12.2 | 12.0 |
| CFE-CGC | 5.8 | 5.8 | 6.1 | 6.0 |
| Other unions | 7.3 | 7.0 | 6.8 | 6.4 |
| Non-union candidates | 25.1 | 24.1 | 27.4 | 26.9 |
CGT = General Confederation of Labour ( Confédération générale du travail ); CFDT = French Democratic Confederation of Labour ( Confédération française démocratique du travail ); CFTC = French Christian Workers' Confederation ( Confédération française des travailleurs chrétiens ); CGT-FO = General Confederation of Labour-Force ouvrière ( Confédération générale du travail - Force ouvrière ), CFE-CGC = French Confederation of Professional and Managerial Staff-General Confederation of Professional and Managerial Staff ( Confédération française de l'encadrement - Confédération générale des cadres ).
Source: DARES (after corrections)
As indicated by table 1, the various trade union confederations either improved their share of the vote from 1996 to 1998 (CGT, CFDT and CFTC), or remained at their previous level (CGT-FO and CFE-CGC). Losses were shared by "other unions" and independent candidates. However, the gains made by confederations were relatively low, especially from a long-term perspective - as indicated by the figure below
Source: DARES.
The gains and losses in votes are greater when viewed from regional and sectoral perspectives. CGT increased its share of the vote in 10 out of 16 sectors, as did CFDT. The former lost ground in four sectors, while the latter did so in five. However, CGT increased its vote in 16 of the 23 regions, whereas CFDT made ground in only nine, and lost votes in 13. Other unions and independent candidates lost votes in 13 sectors and 11 regions, and 10 sectors and 6 regions respectively.
The data currently available do not allow for an assessment of whether the drop in votes for independent candidates is due to the unionisation of formerly non-union works councils, or changes in voting behaviour during elections in which independent slates were competing with union slates. Independents lost ground both in councils elected by a single electoral college (where their vote fell from 45% to 43.4%) and in works councils elected by a number of colleges (works council elections may be conducted through separate colleges for blue- and white-collar workers and managerial and professional staff). This is borne out by the fact that independents lost votes in companies of all sizes up to those with 1,000 employees. However, their vote in firms with over 1,000 employees rose, albeit still attaining only very low levels of support (3.3% in 1998 compared with 2.5% in 1996).
Independents still largest single group
Despite their losses, independents still constitute the largest single group of works council members. Their predominance in small workplaces has enabled them to convert their overall vote into a much higher proportion of all works council members: around 43% of all works council members elected in 1998 were independent candidates, though they won only 24.1% of all votes cast. The two largest unions, CGT and CFDT, have a combined total of only 36% of members but won 46% of the vote. CFDT, better supported in smaller workplaces than the CGT, won fewer votes overall than the latter, but had more members elected.
| 1998 | 1997 | |
|---|---|---|
| Total no. of works council members | 59,118 | 57657 |
| Proportion of works council members elected: | . | . |
| CGT | 17.6% | 16.0% |
| CFDT | 18.2% | 17.2% |
| CFTC | 3.5% | 3.6% |
| CGT-FO | 9.7% | 9.1% |
| CFE-CGC | 4.5% | 4.5% |
| Other unions | 3.8% | 3.5% |
| Non-union candidates | 42.8% | 46.0% |
| Total | 100% | 100% |
Source: DARES (unpublished).
Commentary
The differences in results in the 1996 and 1998 elections, in terms of both turn-out and the performance of the various participants, make sense only when viewed in the context of long-term trends. The small degree of difference between the results requires a careful interpretation of them, all the more so because of problems in gathering the data. These problems have meant changes in the area covered by the research, because the companies surveyed are not the same ones as before. The Ministry of Employment and Solidarity has made greater than usual statistical adjustments: in the départementof Paris, one-third of the results were not submitted, and had to be pieced together again separately. The uncertainty resulting from these changes may be a more significant factor in explaining the differences in the results than the performances of the actors themselves. The results can thus only be treated as provisional data. This will also be true of those for the elections held in 1999 and 2000.
The results in private sector works council elections are important because they measure support for trade unions, which often make reference to them when estimating their respective strength. The election results are particularly significant because unionisation figures are based solely on membership figures announced by the unions themselves. The methods the unions use for counting their membership vary widely and are considered unreliable. The issue of the representativeness of unions, the principles on which this concept is based and the use made of it, as well as the measurement of this representativeness, are currently subjects of serious discussion (FR0006170F). Unions require representative status in order to enjoy certain privileges, including being able to negotiate sector-level or company agreements which are binding on all the employees covered by such agreements (even if they are the only union signing the agreement). The unions affiliated to the five confederations with nationally representative status - CFDT, CFE-CGC, CFTC, CGT and CGT-FO - are exempt from having to prove their representativeness at sector and company level, while independent union organisations have to do so.
The unions, the government, employers and the media have engaged in a polemical debate on the representativeness issue. The idea of "majority unions", measured by the combined electoral support for the unions in question, is often opposed to the old principle of representative status. According to one theory being mooted, collective agreements should no longer be valid unless signed by the unions which have the combined majority of employees' votes in the relevant bargaining unit (sector, company etc). In this context, the works council election results would gain in importance for the assessment of the respective strength of support for the various unions.
The works council elections have highlighted some of the difficulties linked in using them for the purpose of measuring support for unions, which include the following:
works councils cover around 5.5 million of the 14 million private sector employees. The number of voters involved is thus smaller than that for the election of workforce delegate s (délégués du personnel), who are elected in all companies with more than 10 employees, not only those with at least 50 as for works councils. Yet the role of workforce delegates, whose task is the daily monitoring of employees' working conditions, is not as prestigious as that of the works council, which is responsible for tasks which are considered to be more "noble" (economic matters and social and cultural activities);
there is a problem concerning the timescale on which figures are collected. The continuous nature of works council elections, and the way in which statistics are gathered, mean that the results of elections that took place in January 1998 are not available until late 2000, 33 months later. Furthermore, they make sense only when added to the data from the previous year; and
there is a problem of the quality of the data-gathering. The statistics drawn up by the Ministry of Employment and Solidarity do not represent the full record of all the elections that took place, because some of the results are not handed in to the Ministry. This lack of exhaustiveness should be allowed for in assessing the results' credibility, particularly when changes in results are very small. Moreover, the results do not just present a picture of the elections that actually took place, but are also the result of technical operations which piece together the voting patterns using formulae for probability. These data are satisfactory for academic use, but can they lay claim to similar authority if they are used to legitimise the results of collective bargaining? The exhaustiveness, credibility and continuity of the data-collection chain are essential. (Christian Dufour, IRES)
Eurofound recommends citing this publication in the following way.
Eurofound (2000), 1998 works council election results published, article.
This website is an official website of the European Union.
All official European Union website addresses are in the europa.eu domain.
See all EU institutions and bodies