Using data from the CPS this paper examines the role of birth-country networks on immigrants' unemployment duration from 2001 to 2013. We find that networks significantly lower unemployment duration for all immigrants. Varying the effect of networks over duration categories we find that networks are more effective in lowering duration for immigrants unemployed for 1-2 months than immigrants who are unemployed for longer periods and this effect is further strengthened during the post recession period. This supports the Calvo-Armengol and Jackson hypothesis which posits that longer the agent is unemployed, less effective are her social networks in job search. Our findings are robust to different specifications.