Macroeconomics textbooks often have what I’ll call “potted histories” of various fiscal and monetary shocks. But they are typically missing the corresponding “potted histories” of technology shocks—especially sharp technology shocks that would cause fluctuations, as distinct from graduate, smooth technological improvement.

I view macroeconomic technology shocks in this way. When there is a new technique, it takes some time to get adopted. The adoption curve is typically S-shaped (similar to a logistic curve). I see the macroeconomic technology shock as a reflection of the steep part of the S-curve where adoption is proceeding quickly.

This view of technology shocks has an important consequence: it predicts that macroeconomic technology shocks can be predicted by identifying techniques in the early adoption phase, before the steep part of the S-curve is reached. There is a difficulty: it is not easy to predict what the final level of penetration of the technique will be, and thus hard to predict whether there is a bigger surge of adoptions yet to come, or whether adoption will start slowing down. But the real possibility of a substantial macroeconomic technology shock will be evident when a technique reaches a certain level of early adoption.

An important part of good monetary policy is responding to technology improvements with expansionary monetary policy that helps output quickly move up to the new higher natural level of output, closing the output gap. In “Are Technology Improvements Contractionary?” and its sequel “Sector-Specific Technical Change” (with Jonas Fisher as well), Susanto Basu, John Fernald and I find that, historically, the Fed has not done a good job with this. Part of the difficulty is conceptual—to realize that closing the output gap doesn’t always mean trying to stabilize output—but the other part of the difficulty is detecting technology improvements early on. The S-curve perspective can help. Central banks with different native languages could collaborate to monitor publications in their own languages that give early warnings about technological advances.

I’d love to have a bigger collection of technology shock stories. Containerized shipping is one. It makes a good story. But I’m not clear whether it had a fast-enough-moving adoption phase to have generated a substantial macroeconomic technology shock.