We investigate the role of information spillovers (IS) in determining firms' labour adjustments. We test the proposition that information on relevant state variables spills over through one firm's decision to affect those of other firms. Our test is based on the assumption that spillovers matter only among firms that are both similar and geographically close. Using a large panel of manufacturing firms, we identify those that are located in a given industrial district and produce the same goods as satisfying both criteria. We propose a solution to the identification problem typical of the empirical analysis of social effects. Our results show that firms' decisions are indeed affected by those of similar, neighbouring firms, while the actions of firms not satisfying either of the criteria have no impact. We test other implications of the theory and find further supporting evidence of the relevance of IS. First, measures of extreme adjustments exert a stronger influence than mean adjustments; second, smaller firms seem to rely more on external sources of information; third, the effects depend on a number of the reference group's characteristics, such as its size and the presence of large firms. Finally, given that firms exposed to IS tend to adjust simultaneously, we find that spillovers amplify the effect of aggregate shocks and constitute a powerful mechanism of amplification of the business cycle.
