WEBINAR RECORDING · 2022-11
Can Evidence Based Information Shift Preferences Towards Trade Policy?
In this conversation: Maggie Chen (Presenter); Lucian Cernat (Discussant); Ben Shepherd and Hanna Norberg (Moderators)
Randomized US survey experiments show that accessible research findings need not move trade preferences in the expected direction. Price-benefit messages can increase protectionist preferences. Discussion separates these experimental results from promising but untested approaches involving locally relevant information, adjustment assistance and participation.
Session information draws on the recording and its original description. Affiliations and policy context refer to the time of the webinar.
FROM THE CONVERSATION
Three takeaways.
What they mean in practice.
Prepared from the session recording using automatic transcription and editorial review. Follow the timestamp links for the original wording and context.
Benefits information can provoke protectionist responses
Watch 23:52 ↗Manufacturing-job-loss information increases support for import restrictions. Service-job-gain messages do not generate a symmetric liberalizing response; lower-price and tariff-cost narratives can also increase protectionist preferences. Similar patterns recur across 2018–22 rounds.
Policy interpretation: Test audiences' interpretations instead of assuming evidence of gains automatically increases support for openness.
Preference strength depends on the question
Watch 20:05 ↗Over half support more import limits in a binary question, but restrictions rank lower when respondents choose among policy instruments. Minimum wages and progressive taxation attract stronger relative preference. Endorsing a proposition is different from prioritizing it.
Policy interpretation: Ask about alternatives and tradeoffs when measuring demand for protection.
Priors and attention shape interpretation
Watch 28:51 ↗Measured economic exposure shows limited treatment heterogeneity, while prior beliefs, loss aversion and political identity show more traction. Respondents connect prices to jobs and China. Longer attention is associated with more message-aligned responses, but attention is not randomized.
Policy interpretation: Address concrete job concerns and test formats permitting reflection; do not treat these associations as proven remedies.
From the discussion
Cernat describes geolocated firm information in CETA communication and favorable feedback, while acknowledging he cannot quantify changed minds (44:42–49:33). Chen treats personalized real-world information as a future experimental direction (55:30). Asked about exclusion from policymaking, she clarifies that trust was measured but participation was not (58:08). Another question distinguishes worker and consumer identities in interpreting lower prices (59:55).
