TRV-2026-3078Version 1 · Published

Saved · see the current record

Why this version exists

Certified into the record

The exact text we saved

This is the version exactly as it was saved, character for character. The SHA-256 hash below is a 64-character code worked out from that text. Run the same text through SHA-256 and you get the same code; change a single letter and the code comes out completely different. Each version also stores the previous version’s hash, which chains them together so no version can be swapped out quietly.

TRUVACE RECORD VERSION
record: TRV-2026-3078
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-11-20T18:02:03.411435Z
status: published
lens: p_space
sector: labor
headline: The impact of fear of automation
dek: In this paper, we study the causal effect of workers’ perceived probability of job loss due to automation on policy preferences and workplace intentions. Using a representative sample of the U.S. workforce, we elicit individuals’ perceived likelihood of losing their job to robots or artificial intelligence within the next decade. We document strong cross-sectional relationships between automation fear and support for higher taxation, more generous government transfers, union membership, retraining, occupational…
gain_title: (none)
problem_title: In a randomized survey experiment with a representative sample of US workers, learning that their occupation faced worse automation job-loss odds than expected raised demand for redistribution, government support and unionization, but left intentions to retrain or switch occupations largely unchanged.
trace_subject: (none)
gain_reading: (none)
gain_evidence: (none)
problem_reading: In a randomized survey experiment with a representative sample of US workers, learning that their occupation faced worse automation job-loss odds than expected raised demand for redistribution, government support and unionization, but left intentions to retrain or switch occupations largely unchanged.
problem_evidence: Adverse information increases demand for redistribution, government support programs, and unionization. | In contrast, it has limited effects on intentions to retrain or switch occupations.
quick_read: Economists ran a randomized information experiment on a representative sample of the US workforce. Workers first stated how likely they thought it was that robots or artificial intelligence would take their job within a decade. Some then received occupation-specific job-loss probabilities, which gave them belief shocks relative to their own priors.

Workers who received worse news than they expected raised their demand for redistribution, government support programs and unionization. Intentions to retrain or switch occupations moved little. Workers facing automation risk appear to look to collective protection rather than individual adjustment.

The effects came mainly from people who got worse news than they expected. The pattern, with shifts after negative shocks and no parallel shifts on related individual choices, makes a simple social-desirability explanation unlikely.
limitation: The study measures expressed policy preferences and stated intentions after an information treatment, using a reduced-form approach to perceived downside exposure, not workers' actual later behaviour or job outcomes.
tag: Evidence-backed problem
key_points: The study elicited workers' perceived likelihood of losing their job to robots or AI within the next decade, using a representative sample of the US workforce. | Before the experiment, fear of automation was strongly correlated with support for higher taxes, more generous government transfers, union membership, retraining, occupational switching, populist attitudes and voting intentions. | A randomized treatment gave workers occupation-specific job-loss probabilities, creating exogenous belief shocks relative to their own priors. | Adverse information increased demand for redistribution, government support programs and unionization. | Effects on intentions to retrain or switch occupations were limited, and the treatment effects were driven mainly by people who received worse news than expected.
rundown: The design separates correlation from cause. The cross-sectional data link automation fear to many attitudes, including populist views and voting intentions. The randomized treatment tests whether changing beliefs about job-loss probability changes those preferences.

The responses were asymmetric: negative belief shocks shifted preferences, while the closely related individual choices (retraining and occupational switching) showed no parallel effects. The authors say this makes a simple social-desirability explanation unlikely.

The authors frame the results around perceived employment risk as discussed in public debate. They conclude that collective protection, not individual adjustment, is the main way workers might respond to automation risk.
sources:
- peer_reviewed | Labour Economics | https://doi.org/10.1016/j.labeco.2026.102979 | 2026-11-04
prev: 0000000000000000000000000000000000000000000000000000000000000000
SHA-256 hash
49f6f97671470c2e508d06a29ecee7c58112f1975440b01a7007512ae52df88c
Previous version’s hash
0000000000000000000000000000000000000000000000000000000000000000
Check this record hasn’t been altered
How to check it yourself, without trusting this page

Download any version from /api/record/TRV-2026-3078 and hash its canonical field yourself, for example with shasum -a 256. You should get its content_hash. Each version’s text also ends with prev:and the hash of the version before it (version 1 points to 64 zeros). If anyone had changed a single character of any version after it was saved, those hashes wouldn’t line up.