Why do some jobs pay so much less than they used to, even after adjusting for inflation?
Background
The unemployment rate is published monthly and cited constantly, but it is an output of a system that is running continuously underneath. Workers are being hired and separated every day. The rate at any moment reflects the balance between those flows.
Search-and-matching theory treats the labor market as a matching problem: firms and workers find each other imperfectly, at cost, over time. That friction is not a market failure to be fixed; it is a structural feature of how labor markets work. The policy question is how much of it is structural versus cyclical -- and that distinction is harder than it sounds.
What it covers
The unemployment rate is the surface reading. Beneath it are flows: workers separating from jobs, workers finding new ones, workers leaving and re-entering the labor force entirely. The steady-state unemployment rate in a search-and-matching framework is simply the separation rate divided by the sum of separation and job-finding rates. When either rate shifts, the equilibrium shifts with it.
The Beveridge curve plots unemployment against job vacancies. When the curve shifts outward -- more vacancies coexisting with more unemployment -- it signals a matching efficiency problem alongside weak demand. That distinction drives very different policy responses.
Open question
Why does unemployment persist even in growing economies, and why does it respond differently to recessions in different countries?