206. Paul Krugman talks trade, industrial policy, and Trump

Episode Summary

Executive Summary: Paul Krugman traces how agglomeration economies made industries cluster geographically, explains why those clusters now create strategic vulnerabilities, and argues for targeted industrial policy rather than tariffs. He says the world is riskier than in the past, making supply-chain diversification and climate-linked subsidies more defensible, while Trump’s tariff strategy is incoherent, destabilizing, and likely to raise costs without fixing trade imbalances.

Main Topics: Agglomeration externalities and geographic concentration (Priority: 5/5): Krugman explains how supplier networks, thick labor markets, and knowledge spillovers create powerful clustering forces that make industries concentrate in a few places. Historical and modern examples of industrial clusters (Priority: 4/5): The conversation reviews classic U.S. and global examples—from Troy collars to Silicon Valley, New York finance, and Chinese manufacturing towns—to show how concentrated production emerges. From efficiency to security: the new risks of concentration (Priority: 5/5): Krugman says his earlier work focused on benefits, but today policymakers worry about pandemics, geopolitics, climate shocks, and financial leverage from concentrated production and finance. Industrial policy, CHIPS, IRA, and strategic diversification (Priority: 5/5): He argues subsidies are often better than tariffs for building domestic or allied capacity in strategic sectors, and says the CHIPS Act and IRA reflect both security and political-economy realities. Europe, Draghi, and the productivity gap (Priority: 3/5): Krugman frames Europe’s relative weakness as partly a consequence of missing U.S.-style tech and finance clusters, while noting the global economy may benefit from those concentrations. Trump tariffs and the incoherence of tariff policy (Priority: 5/5): Krugman criticizes Trump’s focus on bilateral deficits, trade deficits, and unstable tariffs, arguing they raise costs, distort investment, and lack coherent objectives.

Key Arguments: Agglomeration economies are real and persistent: suppliers, labor-market matching, and spillovers make production cluster in a few locations. Krugman says early economists emphasized the upside of clustering but largely ignored downside risk because the world seemed less dangerous. Concentration now matters more because supply chains can be disrupted by geopolitics, sanctions, pandemics, climate shocks, and trade wars. Industrial policy is preferable to tariffs when the goal is to build capacity in strategically important sectors, because tariffs raise costs for downstream users. The CHIPS Act and IRA are defended as responses to national-security needs and to the political difficulty of using direct taxes like a carbon tax. A key advantage of subsidies is that they can be politically durable once firms and communities become invested in them. Industrial policy can still waste money and widen capture risks if it expands beyond strategic targets, such as attempts to reshore nonstrategic goods. Trump’s tariff logic is incoherent because bilateral deficits are not a meaningful target and the overall trade deficit is driven by capital flows, not tariffs. Tariffs generally strengthen the dollar and reduce both imports and exports, so they are unlikely to fix the trade deficit. Uncertain, on-off tariffs are especially damaging because they paralyze firms’ investment planning. Krugman says his 1987 skepticism about industrial policy was partly wrong because he underestimated security concerns and political uses of subsidy coalitions. He warns that scrapping industrial policy while using tariffs would protect low-value sectors while sacrificing high-value strategic manufacturing.

Data Points: Years on NYT column: 25 years - Introductory description of Paul Krugman’s career Nobel Prize year: 2008 - Krugman won the Nobel Prize for Economics New York metropolitan area population: 23 million - Used to illustrate agglomeration in a dense metro region U.S. semiconductor concentration in Taiwan: More than 90% - High-end semiconductors are described as being manufactured in one location on Taiwan U.S.-European productivity differential attributed to clusters: Roughly half - Krugman estimates about half of the gap comes from Silicon Valley, Seattle, and New York agglomerations U.S. prime-age employment in Europe comparison: Higher than the U.S. - He notes Europe has more prime-age workers employed than the U.S. Life expectancy gap: Several years longer - Krugman says Europeans live several years longer than Americans U.S. unemployment rate: 4% - Used to argue the U.S. trade deficit is not inherently a problem in a full-employment economy Time span of free trade agreement with Canada: 30 years - Krugman notes the long-standing U.S.-Canada free trade relationship Time since classic industrial policy debate: Late 1980s / early 1990s - He refers to the earlier U.S. industrial policy debate when he opposed it Tariff level example: 20% to 25% - European value-added taxes mentioned as a potential mistaken target for tariff-style retaliation Timeline of CHIPS Act and IRA: A little over two years - He says lessons from these policies are still early and tentative

Pivotal Quotes: "I devoted approximately zero time to worrying about the risks." — Paul Krugman: Reflecting on his early research era, when agglomeration benefits seemed obvious and geopolitical risk seemed remote "The whole logic of agglomeration says that there's a cumulative process where an agglomeration gets established and it's very hard to dislodge later on." — Paul Krugman: Explaining why strategic diversification is difficult but sometimes necessary "The world became more dangerous, which was one reason that industrial policy became suddenly popular again." — Paul Krugman: Summarizing why subsidy-based industrial policy has regained support

Implications: Listeners should expect more state-led efforts to diversify supply chains, protect strategic industries, and subsidize clean-tech production. Tariffs are likely to keep raising costs and creating uncertainty, while industrial policy will remain politically attractive but must avoid waste and capture.