This paper uses two recent UK surveys to investigate labour market performance, the determinants of language proficiency, and the effect of language on earnings and employment probabilities of non-white immigrants. Our results show that language acquisition, employment probabilities, as well as earnings differ widely across non-white immigrants, according to their ethnic origin. Language has a strong and positive effects on employment probabilities. Furthermore, lack of English fluency leads to substantial earnings losses of immigrants. While earnings of white and ethnic minority natives develop in a similar manner, there is a large earnings gap between these two groups, and ethnic minority immigrants. English fluency contributes considera-bly to reducing these differences. Addressing the problems of measurement error and unobserved heterogeneity in language variables, our results indicate that measurement error in the language variable leads to underestimation of the importance of language for employment probabilities and earnings in straightforward regressions. In comparison with results found for other countries, language proficiency seems to be more important for labour market outcomes of UK immigrants.