Even when benefits protect them, those who lost their jobs fall behind
In all 42 matched comparisons in the automation simulation, displaced workers ended below the path in which the change did not happen, even when benefits maintained their level of consumption.
Maintaining today's level of consumption is not the same as not losing out. In all 42 matched comparisons, displaced workers ended below the control path, while workers who remained employed ended above it. This is a simulation of assumptions, not a forecast.
What did we examine?
We built 576 combinations of assumptions about a world in which automation leads to 20% unemployment for 20 years: how much productivity rises, how much of the savings passes through to prices, what the tax rate on profits is, what share of profits is accessible to taxation at all, what happens to housing and interest rates, and how concentrated capital ownership is. In each combination we followed four groups: workers who remained employed, owners of productive capital, and two groups of displaced workers, with and without consumer debt.
First finding: consumption can be maintained, but it is not guaranteed
202 paths reached the end of the 20 years in all seven financing variants we tested. In all 202 paths, workers and capital owners maintained at least 90% of their initial consumption (excluding housing) in every year. For displaced workers the picture is different: only in 75 paths did both groups stay above this threshold. In other words, falling prices and benefits can protect, but only under certain conditions.
Second finding: maintaining the level is not the same as not losing out
In 42 paths there is also a complete control path, that is, exactly the same world without worker displacement and without the productivity surge. In all 42 comparisons, and in all financing variants, the consumption of both groups of displaced workers at the end of the period was lower than it was on the control path. Workers who remained employed ended above the control in all 42 cases, and capital owners in 30 of them. A benefit that maintains today's standard of living protects against a fall, but not against a gap that keeps widening as the economy grows.
What distinguished good paths from bad ones?
When all profits were accessible to taxation, displaced workers' consumption was maintained in 65 out of 130 paths, compared with 10 out of 72 when only half was accessible. A 35% profit tax yielded 50 out of 96, compared with 25 out of 106 with a 25% tax. Full pass-through of savings to prices lowered preservation mainly among renters and mortgage holders: rent, mortgage payments and maintenance are fixed amounts that do not fall with prices, while the benefit targets purchasing power over goods only. Among renters, preservation fell from 8 of 16 paths without pass-through to 2 of 18 with full pass-through; among outright owners it barely changed (8 of 18). The fiscal room for benefits also shrank, and at the unemployment peak it ran out in most paths, so indexing the benefit to housing costs does not undo the harm to renters either. These are descriptive counts within a model, not economic laws and not causal estimates.
Limitations
All inputs are assumptions, not calibrated estimates. 356 of the 576 combinations did not reach the end of the path in one or more of the financing variants, and we draw no conclusion about them. The control paths keep the policy and the housing schedules unchanged. The model does not claim that automation will cause 20% unemployment, and it does not predict who will actually be affected.
Sources and transparency
This publication explains the existing research in accessible language. Its accounting example and scenarios are neither official forecasts nor investment recommendations.
What do you think?
Should a benefit maintain today's standard of living, or a fixed share of the wealth the economy produces?