This article exploits changes in the distribution of immigrants across 20 Organization for Economic Co-operation and Development countries from 1960 to 2005 in order to assess their contribution to income of destination countries. The non-random sorting of immigrants across countries is addressed by using an instrumental variable strategy. The instrument is built by estimating a bilateral migration model incorporating exogenous origin country determinants of migration. Aggregate results reveal that immigrants have a positive effect on income that works primarily through total factor productivity (TFP). We further construct a novel dataset from censuses and labor force surveys to explore the information on the age of immigrants. Contrasting income effects are found across age groups: a higher share of immigrants among the youth has a negative impact on aggregate income, while a higher share of immigrants among prime-aged workers has a positive effect. We interpret this disparity as short-term versus medium-term effects. Adjustments over time involve changes in TFP but also in the human capital of the native-born. (JEL F22, J24, J31, O31)