How Joshua Angrist might approach Economics

Economics, as a discipline, ought to be more like baseball. Not the statistics alone, mind you, but the *game*. What’s the real question we’re trying to answer? Is it *what* happened, or *why* did it happen? Too often, we get lost in batting averages and stolen bases, forgetting the swing of the bat, the curve of the pitch. My concern isn't with elegant theories that sit on a shelf, but with understanding what makes the economy *work*, what makes people’s lives better or worse.

The trick, of course, is getting at *causality*. Anyone can show you two things moving together. That’s correlation, and frankly, it’s cheap. The real challenge, the gritty work, is figuring out if A actually *causes* B. And the only way to do that credibly is with a solid research design. What’s the source of random variation? Where’s the natural experiment? A draft lottery, a sudden policy change, a cutoff score – that’s the kind of stuff that gives us a real shot at isolating an effect.

Without that, what are we left with? A lot of speculation, a lot of ‘what if’ scenarios that can’t be tested. We need to be able to look at the data and say, "Yes, this is what happened because of *that*." When I see analyses that rely on opaque assumptions, that can’t point to a clear instrument, that’s just not credible. It's not about making the math pretty; it's about making the *story* true. And for that, you need a good identification strategy. The data don't lie, but they do require careful interrogation. Let's look at the first stage.

Imagined perspective — an AI synthesis grounded in Joshua Angrist’s recorded ideas and methods, not a quotation or a statement they actually made.