This paper investigates the effect of minimum wages on employment using a panel of US state-based data. We estimate a minimalist dynamic version of the specification implied by neo-classical theory. We find statistically and economically significant effects of minimum wages on youth employment. Unlike many other studies we find also significant effects on aggregate state employment. These results re-establish the conventional wisdom as existing before the work of Card-Krueger-Katz. The paper meets the methodological criticisms of this sort of panel study made by CKK. An important econometric innovation in this paper is to produce estimates allowing for cross-sectional correlation, which offers unbiased inference and potential efficiency gains.