Purpose The purpose of this paper is to study the implementation of a new open-ended contract termination in 2008 in France, called the rupture conventionnelle (RC), which is a mutually agreed contract termination. More precisely, this paper analyses first the impact of the RC on the employers’ termination decisions (termination or not?). Then it seeks to provide empirical evidence of a substitution between the RC and other contract terminations (if there is termination, what types?). Design/methodology/approach The approach is first empirical. The author uses two matched firms’ data sets: one relating to the movement of employees and other from accounting data from 2006 to 2009. Using a propensity score matching method, the author creates two similar (from observable characteristics) firms’ groups – those that used RC in 2009 and those not. The author compares the evolution of the employment decisions between the two groups between 2008 and 2009 in order to identify the specific effect of the RC in the user firms. Findings The results indicate that the introduction of the RC tends to increase workforce exits and does lead to much more job destruction in the user firms, i.e. job destruction would have been weaker in the user firms if the RC had not been implemented. Substitutions with dismissals for personal reasons and for economic reasons also may have appeared, albeit weakly. Originality/value The RC seems an important measure to make the labour market more flexible, especially in France where EPL is often criticize for its rigidities. But few studies analyse the consequences of this new institutional rule on firms’ behaviour and with firms’ data set. Then this paper provides first some understanding and assessment of the economic effects of the RC.