In many countries, notably across Europe, collective bargaining coverage is enhanced by government-issued extensions that widen the reach of collective agreements beyond their signatory parties to all firms and workers in the same sector. This paper analyses the causal impact of such extensions on employment using a natural experiment in Portugal: the immediate suspension by the government that took office in 21 June 2011 of the (until then) nearly automatic extensions. The combination of this suspension and the time needed for processing the extension applications resulted in a sharp and unanticipated decline in the extension probability of agreements signed several months earlier, around 1 March 2011. Our results, based on a regression discontinuity design and matched employer-employee-agreement panel data, suggest that extensions had a negative impact on employment growth. Moreover, the effects tend to be concentrated amongst non-affiliated firms. The lack of representativeness of employer associations is a potentially important factor behind the adverse effect of extensions. Another is the role of retro-activity in combination with the administrative delay in processing extensions. This is particularly relevant in the context of a recession.