True, but to be fair a null result is much more interesting (and thus publishable) when it cuts against a large body of prior empirical work or theory. Thinkin

True, but to be fair a null result is much more interesting (and thus publishable) when it cuts against a large body of prior empirical work or theory. Thinking specifically of Card and Krueger (1994) finding zero impact of the minimum wage on employment