This paper explores whether the state provision of school meals in the 1980s crowded out private provision by examining two policy reforms that radically altered the UK school meal service. Both reforms effectively increased the cost of school meals for one group (the treated), leaving another unaffected (the controls). I find strong evidence of crowd out: the reforms reduced school meal take-up among the treated by 20–30 percentage points, with no difference among the controls. I then examine whether this affected children's body weights, using a large, unique, longitudinal dataset of primary school children from 1972 to 1994. The findings show no evidence of any effects on child body weight.