Greg Ip on Industrial Policy

Episode Summary

Executive Summary: The episode examines U.S. industrial policy through the lens of semiconductors, energy, and other strategic sectors. Greg Gipp argues that government intervention works best when it induces broadly valuable innovation for clear national-security or public-good reasons, but fails when it becomes protectionism, over-specified technology picking, or a vehicle for rent-seeking and political strings.

Main Topics: What industrial policy is and when it is justified (Priority: 5/5): Gipp defines industrial policy as government favoring specific industries in the national interest and distinguishes between broad support for R&D/public goods and narrow support for national champions or protectionism. Historical examples: tariffs, defense, and induced demand (Priority: 5/5): The discussion traces U.S. industrial policy from Hamilton’s tariff-based manufacturing strategy to Cold War/space-era demand that helped seed the semiconductor industry. Mixed record of renewable energy industrial policy (Priority: 4/5): The conversation contrasts failed technology picks like Solyndra and the Clinch River breeder reactor with broader incentives that accelerated renewable deployment and helped firms like Tesla, even if much manufacturing moved abroad. CHIPS Act as targeted industrial policy (Priority: 5/5): Semiconductors are treated as a special case because of national security, supply-chain fragility, and concentration of manufacturing in Taiwan, South Korea, and China; the panel debates whether the CHIPS Act is sufficiently focused. Conditions, rent-seeking, and unintended consequences (Priority: 4/5): A major theme is whether attaching labor, childcare, buyback, profit-sharing, and China-investment restrictions undermines take-up and commercial viability. China, strategic competition, and the return of industrial policy (Priority: 5/5): The rise of China is presented as the key reason the free-market/neoliberal consensus weakened and why the U.S. is increasingly willing to use industrial policy to reduce dependence on rivals. How success should be measured (Priority: 4/5): Success is framed as increasing U.S. fabrication share, building advanced-node fabs, creating clusters, and ensuring projects can survive once subsidies end.

Key Arguments: Industrial policy should be judged by both its goals and its methods: broad public-good support is different from picking winners or protecting declining industries. Government is most defensible when it creates induced demand for products the private sector would not otherwise fund, such as semiconductors, vaccines, or space/missile technologies. Many programs labeled industrial policy are really protectionism, as illustrated by the Jones Act and long-running support for industries with weak competitive advantage. Energy policy is a recurring failure zone because politicians often specify particular technologies rather than setting broad goals, leading to misallocation and dead-end investments. Broader incentives can still work even when individual bets fail: renewable energy subsidies accelerated R&D and adoption, but did not fully build a domestic panel industry. The CHIPS Act is more justified than most industrial policies because chips are foundational to civilian and military systems and manufacturing is concentrated in geopolitically risky locations. The CHIPS Act’s design is relatively better than older programs because it is open to friendly-country firms and seeks clusters and commercial viability rather than forcing every district to get a fab. Adding too many social-policy conditions can deter strong firms from participating, lowering program quality and risking that only weaker applicants seek the money. Industrial-policy waves tend to rise when confidence in market allocation falls; today’s wave is driven largely by skepticism about globalization and competition with China. Industrial policy can be justified by national security or economic security, but the bar should be much higher for sectors like EV batteries or copper, where protectionist lobbying may dominate.

Data Points: Date of episode: Wednesday, March 8th - Opening introduction by the hosts Jones Act year: 1921 - Referenced as the law requiring ships moving between U.S. ports to be built in the U.S. Period of U.S. tariff dominance: Roughly the first century of the republic - Describes the long era in which the U.S. relied on high tariffs Timeframe of postwar industrial-policy model: Last 50 or 75 years - Used to describe the modern conception of industrial policy Semiconductor fabrication share of the U.S.: 12% - Current U.S. share of global semiconductor fabrication, cited as a baseline to improve upon CHIPS Act funding: $39 billion - Amount appropriated for semiconductor subsidies Projected industry investment: $3 trillion - Estimated semiconductor industry investment over the next decade Tariff on imported pickup trucks: 25% - Lyndon Johnson-era tariff mentioned as the 'chicken tax'

Pivotal Quotes: "What we mean or what economists mean when they use the phrase industrial policy is when the government uses its resources or its authority to favor particular industries because they believe that doing so is in the national interest." — Greg Gipp: Opening definition of industrial policy "I think the rise of China changes everything." — Greg Gipp: Explaining why industrial policy has regained support in the U.S. "You do not want a situation where the most financially strong and successful semiconductor chip companies say, you know, frankly, we don't need the money that badly. We're going to go invest somewhere else." — Greg Gipp: Warning that too many strings on CHIPS Act money could deter the best firms

Implications: Listeners should expect industrial policy to remain prominent, but the policy challenge is to target genuinely strategic sectors, avoid overconditioning subsidies, and prevent protectionism from overwhelming national-security goals.