SCENARIO ENGINE · ALL ASSUMPTIONS ADJUSTABLE

The model

The timeline on every city page is a scenario, not a forecast. This page is the entire model: one equation, six parameters, three citable presets. Change anything — the URL encodes your scenario, so you can publish it.

displaced(occ, y) = βocc · a · C · σ(k(y − TL))

βocc = share of the occupation's tasks that are LLM-exposed (Eloundou et al. 2024); σ = logistic. Layoffs = displacement flow above the attrition rate r. Aggregated over each metro's occupation mix.

PRESETS

Year AI systems can perform exposed white-collar tasks at replacement level. The site's main slider.

Logistic steepness per year. 0.8 means adoption goes 12%->88% over ~5.5 years around the midpoint, consistent with enterprise software diffusion (faster than electrification, slower than smartphones).

Maximum share of technically-automatable task value that ever gets automated (regulation, integration costs, human-preference niches).

Share of AI task usage that substitutes for the worker rather than augmenting them. Default from the Anthropic Economic Index automation/augmentation conversation split.

Years between capability arrival and employment effect (contracts, reorgs, attrition-first policies).

Annual displacement FLOW (fraction of an occupation's jobs per year) that retirements/quits absorb without layoffs; only flow above this rate becomes involuntary layoffs. BLS total separations run far higher, but only non-backfilled exits count. Gross = positions eliminated; net = involuntary layoffs.

Figure 1. Black: positions eliminated (% of metro employment). Red: cumulative involuntary layoffs — displacement flow in excess of attrition. Grounding constraint: with any defensible parameters, modeled displacement through 2026 stays under 1.5%, consistent with the observed aggregate null (methods §5).