Trade Uncertainty May Cut U.S. Industrial Output by 1.1%

Trade uncertainty has surged to unprecedented levels, with the potential to significantly impact U.S. industrial output. According to a model prediction, the escalation in uncertainty could lead to a 1.1% reduction in U.S. industrial output by the second quarter of next year, compared to baseline expectations. This impact is substantial, as it represents a nearly 1.7% deviation from the global economic baseline.

The uncertainty surrounding trade policies has led to a halt in investment plans by some enterprises. This hesitation is due to the lack of clarity regarding the implementation details of additional tariffs. As a result, companies are rushing to import goods ahead of potential interest rate hikes, driving the ISM manufacturing inventory index to its highest level since October 2022.

Economists have warned that the new tariffs could lead to a slowdown in economic growth and increased inflation in the short term. Other nations may also retaliate with their own tariffs in response to the White House's actions. The escalating trade war has already taken a toll on the global economy, with the potential for further disruptions looming on the horizon.

The impact of trade uncertainty is not limited to the U.S. economy. The model predicts that global production could also experience a similar decline, with a potential reduction of 1.7% by April 2026 compared to the baseline scenario. This is because countries like Germany and South Korea, which have economies more dependent on trade, are likely to be more affected than relatively closed economies like the U.S. However, these estimates come with a high degree of uncertainty, and the confidence intervals suggest that the possibility of no impact is unlikely.

Historical data suggests that during the first term of the Trump administration, actual tariffs were imposed only after uncertainty had reached very high levels. This indicates that the predictions may be an optimistic estimate of the impact of uncertainty. However, the model also assumes that uncertainty will return to its mean in the coming months, and its impact will gradually diminish. If uncertainty remains high, the economic impact could be more severe.

The analysis is based on a Bayesian VAR model estimated using a trade policy uncertainty index derived from news content. For the U.S. model, variables included the MSCI global index, 2-year Treasury yield, U.S. industrial output, imports, exports, and the real dollar exchange rate. For the global model, the first two variables were retained, and global industrial production and global trade data were included.

The model was estimated using monthly data from 2000 to 2024, and the "pre-pandemic" method proposed by Cascaldi-Garcia (2022) was used to explain the volatility during the COVID-19 pandemic. The trade policy uncertainty shock was determined through recursive ordering, with the trade policy uncertainty index placed last. Confidence intervals were calculated through 5,000 model simulations drawn from the posterior distribution. The responses plotted were calculated in response to the structural trade policy uncertainty shocks observed in November 2024, January 2025, and March 2025.