Haishi Li is an assistant professor in economics at the HKU Business School of the University of Hong Kong. He is also a research affiliate at CESifo. He works on international trade and international macroeconomics, with research interests spanning multinational production, trade policy, emerging technology, and climate change. He has published in leading academic journals including Economic Journal, Journal of Monetary Economics, Journal of International Economics, and Review of Economics and Statistics. He received a PhD in economics from University of Chicago and was a postdoctoral fellow at School of Global Policy and Strategy, University of California, San Diego.

We document new empirical facts on how semiconductor innovation affects downstream innovation in the United States and China. Motivated by these facts, we develop and estimate a two-country endogenous growth model in which semiconductor firms innovate strategically and downstream firms use semiconductors as essential inputs into innovation. We use the model to quantify the effects of U.S. semiconductor export controls. Export controls raise innovation in U.S. semiconductor and downstream sectors in the short run, but reduce innovation in both sectors in the long run, regardless of whether the United States widens its technological lead over China. The short-run gains reflect greater domestic semiconductor supply and stronger escape-competition incentives, while the long-run losses arise from reduced access to the Chinese market. Export controls also narrow the U.S. downstream trade deficit, and AI-driven downstream productivity growth amplifies short-run gains while mitigating long-run losses.