Does private school competition harm public schools? Revisiting Smith and Meier's the case against school choice

Smith and Meier (1995a) empirically assess the market hypothesis advanced b y Chubb and Moe (1988, 1990), which holds that competition improves schools . Using Florida school district data, Smith and Meier find that higher priv ate school market share lowers public school test scores; they conclude tha t competition harms public schools. However, they do not take into account the impact of family income on competition in traditionally organized educa tion markets: the lower the income, the less likely parents can exit the pu blic schools, which implies less competition. Using their database but segm enting it into low and high income groups, we reanalyze the relationship be tween test scores and private school market share. We find that private sch ool market share lowers test scores primarily in low income districts, wher e competition is least due to low family income. We conclude that Smith and Meier's rejection of the market hypothesis is premature.