Incidence of work refusal due to health and safety risks

The French Office for Research and Statistics has looked at how often workers stop or refuse to carry out a job because they believe it may be hazardous or put their health at risk. Workers in France are legally entitled refuse a task for health and safety reasons. While the study found that 12% of all respondents had stopped or refused to carry out a dangerous task at some point, there were wide variations in such behaviour according to occupation, sector, and type of workplace.


For more than 30 years, the French Labour Code has allowed employees to refuse to carry out work if they have reason to believe it presents a serious and imminent risk to their life or health. This right is set out in Article 4131–1 of the French Labour Code and was enforced through the fourth Auroux Law from December 1982. This legislation also obliged companies to set up a health and safety committee (CHSCT) for each workplace.

Similar regulations are in place for the public sector (Decree 82–453 of 28 May 1982). The core of the provision gives employees a right to ‘insubordination’ within strictly defined boundaries. These boundaries include the reasonable belief that their life and/or health are at risk.

When the use of this right has been assessed, the regulations have been found to have limited impact for two main reasons. Firstly, employees knew little about these provisions. Secondly, even those who were well-informed were cautious about make use of the legislation since the law did not give a precise definition of what represented a serious and imminent risk. Employees felt this meant they risked misinterpreting the rules, and this could lead to sanctions and even dismissal (Lachaise, 1991; Chabbi, 2005).

Analysis of the data

In 2010, the Medical Monitoring Survey of Professional Risks Survey (SUMER), included a question on refusing to carry out work for health and safety reasons (FR1301011D). Question 68 the questionnaire (64 KB PDF) asked employees whether or not they had, within the past 12 months, ‘suspended or refused to take up a task in order to maintain their health and safety’.

The data were used to analyse the frequency of health and safety related stoppages at work. The research also examined details about the workplace and the individual characteristics that might increase or lessen the likelihood of this type of incident.

The questionnaire does not, however, make direct reference to the legislation.

Workplace context

Figures show 12% of the employees surveyed reported that they had ‘stopped or refused a task’ for health and safety reasons in the previous year.

The study showed wide variations between socio-professional groups. Blue-collar workers were most likely to refuse tasks (16%), followed by white-collar workers and intermediate professions (11% each), and managers (8%). There were also significant sectoral differences (Figure).

Figures showed women (11%) were slightly less likely to break off from or refuse a task than men (13%).

Figure 1: Proportion of employees stopping or refusing a task by sector

Proportion of employees stopping or refusing a task by sector

Source: Algava et al, 2013

The data suggest a link between the likelihood of an employee stopping a task and the employee’s previous experience of workplace accidents.

Employees who have had at least one accident (17%) are more than twice as likely to refuse a task as those without experience of a workplace accident (7%). Figures also show that the length of absence following an accident is longer for employees who have stopped a task at least once – 23 days on average compared with 18 days for workers who had not broken off from or refused to carry out a job.

The authors of the studies read these figures as evidence that prior experience with workplace accidents increases an employee’s awareness of hazardous situations.

Types of workers most likely to refuse tasks

Using multiple correspondence analysis (MCA), the authors identified seven types of employee who had stopped a task, grouping them according to their working conditions. This approach allowed the mapping of probabilities for employees stopping a task among the different groups.

Table 1 summarises the main characteristics of these groups and the proportion of employees that have stopped or refused to undertake a task.

Worker types most likely to stop/refuse tasks


Work environment

Work autonomy

Gen. health

% female

Typical occupations

% all stoppages

% workforce

% employees in group

Isolated worker

Often exposed to hazardous biological or chemical agents. Often works with clients, rarely with colleagues


Below average


Home care.Cleaning




Harassed worker

Difficult relationship with superior and/or colleagues




All occupations





Atypical working hours. Severe consequences in the case of individual errors. High physical and verbal aggression.

High exposure to biological agents




Health and social services




Crafts worker

Noisy environment and high exposure to chemical agents. Often work in small companies.




Various occupations in industry and construction




Fragile skilled worker

Difficult relationship with superior and/or colleagues


Poor, although the group is characterised by a high share of young workers


Industrial workers in the industrial sector




Stressed worker

Comprehensive responsibilities including over other employees

Very high



Managers and other higher occupations with management responsibilities




Seldom exposed worker

Relatively good working conditions with few occupational risks


Below average


All occupations




Source: Algava et al, 2013

Health issues and collective action

The authors of the study also suggest that poor health contributes to the probability that an employee will break off or refuse to carry out a task. Of those who recorded task stoppages, 29% reported an ‘activity limitation’ and 21% said they suffered from depression or anxiety.

Roughly half of the work stoppages and refusals’ were reported to be part of a collective action. Those surveyed said they had joined colleagues in breaking off from work. It is not entirely clear from the data which decisions to stop work were made by an individual on a ‘right to refuse’ basis, and which were part of industrial action against poor working conditions.

Employee workplace representation

A slight increase in the frequency of work stoppages was discovered in businesses where there was employee workplace representation. In workplaces where a trade union delegate and a workforce delegate were present, 13% of the workforce reported that they had broken off from work or refused a task in the previous 12 months. The proportion dropped to 12% if only one of the bodies was present, and to 10% if there was no employee representative.

A similar finding is reported for companies with a CHSCT (health and safety committee). If a CHSCT is present and if it holds at least four meetings a year, the proportion of stoppages rises to 13%. In workplaces where the committee is less active it drops to 12%, and to 11% where there is no CHSCT.

These results are statistically fairly weak. However, the authors of the study suggest that employee workplace representation bodies help to inform employees about health and safety hazards. They suggest they also let employees know about previous action taken to improve workplace conditions, and about a worker’s right to refuse to carry out a task.


  • (2013), ‘Les salariés déclarant avoir interrompu ou refusé une tâche pour préserver leur santé ou leur sécurité : les enseignements de l’enquête Sumer’ [Employees claiming to have suspended or refused a task to preserve their health or safety: lessons from Sumer survey], Dares Analysis, No. 23, April 2013.
  • D. (2005), ‘Réflexions sur les conditions d’exercice du droit de retrait’ [Reflections on the conditions for exercising the right of withdrawal], Bulletin social, No. 11, November 2005.
  • G. (1991), ‘Le droit de retrait des salariés de leur poste de travail’ [Employees’ right of withdrawal from their jobs], La semaine juridique Entreprises et Affaires, No. 44, October 1991.

Sebastian Schulze-Marmeling, IR Share

Useful? Interesting? Tell us what you think. Hide comments

Eurofound welcomes feedback and updates on this regulation

Tilføj kommentar