An interactive chapter
An apparatus you can move.
Here, a labor supply-and-demand chart — the model both sides of the debate are arguing inside. Drag the floor and watch the prediction respond.
There's no shortage of good learning material. The bottleneck is purpose.
We want answers to the big questions in fields we care about. Are high interest rates good for the economy? Do trade wars help us? Did Keynes mean what the textbook says? To take these seriously you dig past the headline — sometimes through an interactive chapter, sometimes through a traced lineage of each side, sometimes through an empirical atlas. The panels below show all three, on one example.
Here, a labor supply-and-demand chart — the model both sides of the debate are arguing inside. Drag the floor and watch the prediction respond.
Here, the standard view: "in a competitive labor market, raising the wage above equilibrium reduces hiring." Stigler · Friedman · Neumark.
Here, the monopsony view: "real labor markets aren't competitive. A binding floor can pull wages closer to marginal product." Robinson · Card & Krueger · Dube.
Here, state-by-state minimum wages and employment effects. Card & Krueger ran their natural experiment on the NJ–PA border; the empirical fight is conducted on this map.
From Scarcity to DSGE
Played, not just solved
EN · 中文 bilingual
Where the standard model bends
For reading, not exam prep
From scarcity through DSGE — twenty chapters with the apparatus visible.
Built so that the formal models stay manipulable on the page rather than rendered as static curves: shift the demand schedule and watch consumer surplus respond, change the elasticity and the welfare numbers move with you. Each chapter cross-links to the live debates where the model is actually contested — text and debate together give you the apparatus those arguments are conducted in.
Who influenced whom — Smith through Kahneman
Trade, GDP, demographics across 1800 — present
Short, sharp pieces on a current claim or news event — reread through the apparatus the site uses. The provisional layer; where ideas go before they earn a Big Question.
The standard response to the AI-displacement worry runs through the Solow paradox: new technology, by raising productivity, eventually raises real wages — that's the textbook conclusion, sketched in any introductory growth model. The empirical record from 1980 forward is harder to read that way. Median wages decoupled from productivity for thirty years inside the same economy that produced the Solow result.
If your time horizon is a generation, "eventually" is a defensible answer. If your time horizon is a working life, it isn't — and most of the people doing the worrying are working their working life. The model and the lived experience aren't disagreeing about the mechanism; they're disagreeing about what counts as "soon enough."
Read the take →