How did economics learn to make credible causal claims?

Take the structural ambition at its full strength, because it is the right ambition and the rest of the thread only makes sense once you see that. The goal was never to describe correlations. It was to estimate the deep parameters — the preferences, the technologies, the genuine behavioral relationships — that do not change when you change policy. If you actually had those, you could do something no amount of curve-fitting can ever do: evaluate a policy that has never been tried, on a population that has never been treated, by simulating the estimated system forward under the new rule. That is the holy grail of policy economics, and structural estimation is the only approach that even attempts it.

And the assumptions were on the table. A structural model is a complete, stated account of the mechanism: here are the equations, here are the exclusion restrictions, here is what we are assuming about how the economy works. You can argue with it because it commits to something. A reduced-form summary of what happened in the data commits to nothing about what would happen under a different regime. The structural program was the disciplined, honest, ambitious way to do economics — estimate the machine, then ask the machine. The question that would tear it open is not whether the ambition was right. It is whether the program could actually deliver the deep parameters it promised, or only something that looked like them until the policy changed.

Lucas first, at full strength, because the argument is a genuine logical demonstration and not a quibble. The central use of the big models was policy simulation. The thing they held fixed to run the simulation — the estimated coefficients — is exactly the thing a policy change moves, because the coefficients encode behavior optimized against the prevailing rule. There is no patch for this inside the fixed-coefficient framework; the framework’s validity for its own central task is what fails. But here is the calibration that the rest of the thread turns on, and it is the most important single piece of honesty in this story. Lucas was right about the disease and wrong about which cure would win. His proposed remedy was to estimate the truly deep parameters — preferences, technology — that are policy-invariant, and rebuild macro on those microfoundations. The trouble is that those deep parameters are no easier to identify credibly than the structural coefficients they replaced. Identifying a utility function from aggregate data runs into the same wall. That is a large part of why the turn that actually won applied economics was not the rational-expectations-structural turn Lucas wanted. It was something else.

Sims next, and his case is the cleanest statement of the identification disease as a disease of belief. The structural models achieved identification by fiat: they excluded variables from equations not because any theory demanded the exclusion but because the exclusion was needed to make the system solvable. Dress that up however you like; the restriction is still something nobody has a reason to believe. Sims’s “incredible” is precise and damning. His VAR is the honest response — do not claim identification you have not earned; model the joint dynamics, recover what the data actually pins down, and reserve structural interpretation for the few restrictions you can defend out loud.

LaLonde last, and his is the one that drew blood empirically rather than logically. Lucas and Sims argued that the structural program could not credibly identify what it claimed. LaLonde showed it did not. He had something the others did not: a real experiment to check against. When the non-experimental estimators were asked to reproduce the experimental answer, they disagreed with the experiment and with each other. This is the founding negative result the next rung would name as its problem statement: the assumptions the structural estimators depend on are, in LaLonde’s framing, assumptions the data cannot test — and when you find a case where you can test them, against an experiment, they lose.

Argue the design-based turn at full strength, in its own voice, because it is the answer to all three of Stage 2’s critiques at once, and the mapping is exact. It does not simulate fixed deep parameters, so the Lucas critique does not bite: there is nothing being held invariant across a regime change, because the method estimates an effect in a setting, not a structure to be extrapolated. It does not impose incredible restrictions, so Sims’s critique does not bite: the identifying assumption is a single, stated, checkable claim about one source of variation, not a dozen exclusions imposed by convenience. And it reproduces experiments because the entire point is to find variation that mimics an experiment — which turns LaLonde’s demonstration from an indictment into the program’s founding motivation. The thing that killed the structural program is the thing the design-based program is built to deliver: identification you can check.

Joshua Angrist and Jörn-Steffen Pischke named the whole shift the “credibility revolution,” and their claim was not modest: empirical work in economics is far more credible today than it was thirty years ago, because the standard for a causal claim changed. The change was not better technique alone — it was a change in what economists will accept as having shown that one thing causes another. The discipline ratified it with its highest honors: the 2019 Nobel to Banerjee, Duflo, and Kremer for the experimental approach to poverty, and the 2021 Nobel to Card, Angrist, and Imbens for the empirical labor work and the framework that made natural-experiment evidence interpretable.

Labor economics was where the revolution proved itself first and hardest. Card-Krueger on the minimum wage, Angrist on the Vietnam draft lottery, the monopsony program that gave the minimum-wage non-effect a mechanism — this is where the design-based move went from clever trick to field standard, and it is a story with its own lineage worth following on its own terms.

Heckman’s defense, at full strength, because the structural tradition fighting back is the predecessor framework surfaced at its best one last time. The structural program was designed to do something the design-based program structurally cannot: ex-ante policy evaluation — predict the effect of a policy that has not been implemented, on a population that has not been treated, under explicit, stated assumptions. A LATE on the compliers of one natural experiment is silent about a different population, a scaled-up version of the program, or a policy nobody has tried. If those are the questions you need answered — and for most genuinely forward-looking policy design, they are — then you need structure, because only structure even attempts them. Heckman’s “building bridges” argument is not a rejection of the revolution; it is a reminder that the revolution’s win came with a real cost, and the cost is the questions structure was built to answer. The calibration matters: this engagement carries Heckman’s methodological frame, which is right, not his record on specific empirical reanalyses, several of which did not hold up while the design-based findings they contested — Card-Krueger among them — did. The frame survives independently of the contested cases.

And the integration is the thread’s real destination, voiced by the revolution’s own people. Imbens, Athey, Chernozhukov — design-tradition leaders — are now importing structural ambitions back in, disciplined by the checkability standard the revolution established. The synthesis is not “structure was right after all,” and it is not “reduced-form forever.” It is something more precise and more useful: credible identification is the floor, and combining it with structural policy-relevance is the live work. The bridges run both ways now, and the people building them are the ones who tore down the old structure in the first place.