This study aims to identify the causal effects of English proficiency on residential location choices of immigrants. Based on the ideas that (i) immigrants whose mother tongue is linguistically close to English learn the language more easily, and that (ii) young children learn a new language more easily than older children, we construct an instrument for English proficiency, exploiting linguistic distance from English, and age at arrival in the United Kingdom for childhood migrants. Using a unique dataset, we construct various measures of residential clustering aimed at capturing different types of immigrant enclave, and find a negative impact of better English skills on residency in a language enclave, but a positive impact on residency in an ethnic enclave. We also find strong evidence of an impact of poorer English proficiency on living in a neighbourhood of lower quality.