This paper provides an empirical test of the relation between the use of some specific work practices and mutual help among colleagues. Information from a very detailed French matched employer/employee survey is used to construct a measure of cooperation among colleagues and to identify five work practices that, according to the predictions of the theoretical literature, promote cooperation: repeated job interactions, interdependencies in production, job autonomy, and peer monitoring and task variety. The findings suggest a statistically significant, positive and important correlation between job autonomy and mutual help and between peer monitoring and mutual help. However, pairwise combinations of those work practices result to be more strongly associated to mutual help: in particular, interdependencies in production combined with job autonomy and task variety combined with job autonomy. Extrinsic incentives such as team pay and performance evaluation schemes that have actual consequences on salary and career are also positively associated to mutual help. Results on the interplay between explicit and implicit incentives are not statistically significant. Hence the prediction that explicit incentives may crowd out implicit incentives to cooperate is not supported.
