The new econometrics

Lalonde (1986) destroyed the old micro-econometrics as thoroughly as Lucas (1976) destroyed the old macro-econometrics, though it was a more nihilistic destruction (``it just ain't working'') and not a sweet idea as in the Lucas critique. In recent years, these ideas have gained ground, including a symposium in the Journal of Economic Perspectives. Falkenblog has a pithy line: if your result shows up only via coefficient t-stats in 2-stage least squares, but not a graph, it isn't there. As someone who did home-brewn matching back in 1995, I feel pleased about my instincts.