arXiv Science⌕ Search

arXiv · 2610.36171

From Tariff Changes to Trade Effects: A Practical Framework for Empirical Trade Policy Analysis

Abstract

After decades of trade liberalization, the 2020s mark a profound shift in the globalization narrative, with tariffs once again becoming a central instrument of trade policy. This reversal also challenges how the effects of trade policy are evaluated. Much of the empirical literature has been developed to assess bilateral tariff reductions or elimination and the associated erosion of preferential advantages. Rising multilateral tariffs pose a different question: they alter not only the trade costs faced by directly affected exporters but also the relative market access of all competing suppliers, potentially reinforcing existing preferences. This paper develops a practical framework for translating tariff changes into trade effects in this new policy environment. We argue that liberalization and protectionism should not be treated as symmetric policy experiments: their effects operate through different changes in relative market access and therefore require careful consideration of how tariff shocks are measured and incorporated into empirical trade models and counterfactual evaluations.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Maria Cipollina, Luca De Benedictis. 2026-10-26. From Tariff Changes to Trade Effects: A Practical Framework for Empirical Trade Policy Analysis. https://arxiv.org/abs/2610.36171

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

Matching-LLM: Historically Anchored Synthetic Prediction of Public Opinion and Elections

Large language models (LLMs) offer new opportunities for predicting public opinion and electoral behavior, but their synthetic responses may not fully reflect human attitudes and heterogeneity. This study introduces Matching-LLM, a weighted framework that combines LLM-generated responses with historical human behavioral data. Using the World Values Survey, American National Election Studies, and German election survey data, we evaluate its performance in reconstructing survey responses and predicting election outcomes. Results show that Matching-LLM generally improves accuracy relative to LLM-only, including 11 of 12 comparisons on survey questions excluded from calibration, although it does not consistently outperform History-only. In the 2024 U.S. presidential election, Matching-LLM predicts 310 Republican electoral votes, compared with the actual 312, outperforming both single-source benchmarks in electoral-vote accuracy. A supplementary German application extends the framework to a multiparty setting. Overall, the findings highlight the potential of historical data to improve LLM-based predictions while emphasizing the importance of task-specific validation.

econ.GN↗

Narratives to Numbers: Large Language Models and Economic Policy Uncertainty

We study the implications of adopting machine-generated data in economic analysis. Rather than taking the data-generating process as given and correcting estimates econometrically, we evaluate how upstream modeling and researcher choices shape downstream estimates. Revisiting the Economic Policy Uncertainty index, we find that large language models fine-tuned on human audit labels identify nearly twice as many auditor-flagged articles as keyword-based alternatives and are robust to noisy text and linguistic drift. Varying classifier, threshold, and aggregation choices produces substantial differences not only in performance but in what the index captures, reshaping the relationship between policy uncertainty and returns.

econ.GN↗

Leveraging US Commercial Adjustments: MFN Tariffs and Trade Diversion

Most-favored-nation (MFN) tariff increases do more than raise trade costs. By changing the relative tariff treatment of competing exporters, they increase the value of existing trade preferences, reshaping relative market access and trade allocation. While the trade-reducing effects of higher tariffs are well established, much less is known about this mechanism. We investigate whether higher MFN tariffs amplify the competitive advantage of preferential partners, generating trade diversion. Using domestic and international trade data, we estimate a structural gravity model with CES-consistent bilateral tariff margins and examine recent changes in US trade policy as an empirical application. We find that higher MFN protection simultaneously reduces imports and redirects trade toward preferential exporters. These effects are highly heterogeneous across sectors and regions, with the largest losses concentrated in intermediate and capital goods, highlighting the role of MFN tariffs in reshaping relative market access and global trade patterns.

econ.GN↗