The paper studies how social connections affect firm-level hiring decisions and performance. We use register data to characterize the social connections of firms'' employees. For causal identification, we use displacements which create directed supply shocks towards the firms of the displaced workers'' social connections. We make sure that our results are fully driven by these directed supply shocks. Results show that firms appear to prefer employed workers they are connected to over unconnected or unemployed workers when hiring. The employed and connected mostly go to high-productivity firms whereas the unemployed and unconnected tend to go to low-productivity firms. Strong connections - family, recent, durable, formed in small groups, between socially similar agents - matter the most. Displacements shocks cause connected firms, in particular low-productivity ones, to hire those workers they are connected to. Unconnected hires and separations are essentially unaffected. These supply shocks therefore cause the creation of additional jobs which increase firms'' employment. In addition, we use these shocks to show that hiring connected workers has a positive causal impact on firm performance. These results are consistent with a stylized framework where connections reduce hiring frictions and where the firm-level possibility to hire connected workers is a function of changing outside options of these workers.