There has been an increase in the number of children going to school in England who do not speak English as a first language. We investigate whether this has an impact on the educational outcomes of native English speakers at the end of primary school. We show that the negative correlation observed in the raw data is mainly an artefact of selection: non-native speakers are more likely to attend school with disadvantaged native speakers. We attempt to identify a causal impact of changes in the percentage of non-native speakers. Our results suggest zero effect and rule out negative effects.

in the relative size of the group of non-native speakers within the year group (and school) from one year to another. We run some tests to investigate the plausibility of this assumption. We come to the conclusion that this assumption is more plausible in the case of non-white non-native speakers than white non-native speakers. However, our analysis suggests that the bias could be negative for the latter group. This suggests that the 'true effect' is highly unlikely to be negative.When we investigate the shock in the demand for Catholic schooling postenlargement, we find that this strongly predicts the percentage of white non-native speakers in Catholic schools (compared to other schools) whereas it does not predict other key school characteristics (including the percentage of non-white non-native speakers). We use this as the basis of an Instrumental Variable strategy where we analyse how this increase in white non-native speakers influences native English speakers in affected schools. While effects are imprecisely estimated for reading and writing, the estimated effect for maths is positive and significant in some specifications. This small, positive effect for Maths is not generalizable to other schools and contexts. It relates to the group of native speakers for which it is estimated (i.e. those attending Catholic schools with new entrants arriving on account of Eastern European enlargement, who have parents with a preference for Catholic schooling). Given that white non-native speakers are closer to native English speakers in terms of their own maths attainment at age 11 (compared to reading and writing where they are further behind), it makes more sense to find a positive peer effect for this subject. One can only speculate as to the reason. One possibility is that the parents of these children are highly educated and strongly attached to the labour market (as shown in other studies) and this effects the educational progress of their children (making them better peers than the average).Both strategies applied here suggest that negative effects of non-native English speakers on the educational attainment of native English speakers can be ruled out. There is no reason to be worried about the increase in the number of non-native speakers of English in primary schools.

No abstract