Episode Summary
Executive Summary: David Beckworth and Jesus Fernandez-Villaverde discuss the post-pandemic inflation episode, arguing 2021 stimulus likely overshot a demand-constrained economy and that both the Fed and ECB fell behind. They then turn to demographics, warning that global fertility has fallen to replacement or below, implying aging, depopulation, political shifts, and slower innovation. AI may help productivity, but it cannot eliminate the knowledge problem or fully offset demographic headwinds.
Main Topics: Post-pandemic recovery and fiscal/monetary stimulus (Priority: 5/5): Fernandez-Villaverde argues the economy would have rebounded quickly after the COVID shutdown even without as much extra stimulus, though targeted support was needed for households and firms most affected by lockdowns. 2021 inflation and macroeconomic theory (Priority: 5/5): He says 2021 was the pivotal year when policy went too far: demand returned as households sat on excess savings while stimulus remained very large. He frames inflation through a mix of New Keynesian and fiscal theory of the price level, emphasizing the government budget constraint. Europe, energy shocks, and the ECB (Priority: 4/5): He explains European inflation as similar to the U.S. in its demand component but amplified by the energy shock from Russia's invasion of Ukraine. He thinks the ECB also fell behind the curve, though inflation is now easing faster than expected. Global demographic decline and South Korea as a preview (Priority: 5/5): He presents South Korea as a leading indicator of global depopulation, highlighting extremely low fertility and the possibility of massive population loss absent immigration. Economic and political consequences of aging populations (Priority: 5/5): He argues demographic change will reshape labor markets, universities, hospitals, infrastructure, elections, and the balance between secular and religious groups, with older voters and high-fertility communities gaining influence. AI, quantum computing, and growth limits (Priority: 4/5): He is optimistic about computational advances, including a proof-of-concept macro paper on a quantum annealer, but skeptical that AI can solve the knowledge problem or fully compensate for fewer people and harder-to-find ideas.
Key Arguments: The COVID-era output collapse was like an extended shutdown, not a permanent loss of productive capacity; once people returned to work, the economy was likely to revert toward trend without massive extra stimulus. By 2021 the policy mix was too expansionary because demand had recovered while supply remained constrained, excess household savings were high, and government/monetary support was still flooding the economy. Inflation is best understood with a hybrid of New Keynesian and fiscal-theory logic: supply shocks mattered, but expected deficits, low rates, and budget-constraint dynamics also pushed prices up. The government budget constraint is central to macro policy; permanent monetary injections are inseparable from future fiscal adjustments. Europe’s inflation was intensified by its dependence on imported energy, especially natural gas, after Russia’s invasion of Ukraine. Population decline will not be uniform: top cities and elite institutions may remain strong, but lower-tier regions, universities, and campuses will shrink or disappear. GDP per capita will become a less useful headline metric in aging societies because changes in the working-age population can dominate measured output. Demographic decline can alter politics because older people and high-fertility religious groups vote more and become larger shares of the electorate. AI and machine learning cannot solve the knowledge problem because the main challenge is eliciting dispersed, tacit, and incentive-sensitive information, not optimizing once data are known. Even large productivity gains from AI may not show up cleanly in GDP because many of the biggest potential improvements are in services like health, where measured GDP can fall when processes become cheaper.
Data Points: U.S. GDP contraction: about -30% annualized in Q2 2020 - Illustrating the dramatic pandemic shutdown shock U.S. GDP rebound: about +30% to +35% annualized a year later - Showing the sharp recovery after lockdowns American Rescue Plan: $1.9 trillion - Discussed as potentially excessive relative to the estimated output gap in 2021 CBO output gap estimate: about $400 billion - Used by Beckworth to contrast with the scale of fiscal stimulus World population: about 8 billion currently - Baseline for the demographic discussion Peak global population: about 9.5 billion around 2050-2060 - Fernandez-Villaverde's projection before decline begins Global fertility replacement rate: around 2.2 to 2.25 - He argues this is higher than the commonly cited 2.1 because of sex ratios and infant mortality Global fertility rate in 2023: around 2.2 - He says this may be the first year humanity falls below replacement globally South Korea fertility rate: around 0.7 children per woman - Used as an extreme example of demographic collapse South Korea current population: 51.8 million - Starting point for his population decline example South Korea projected population: as low as 12 million - His estimate for the long-run outcome without major immigration China current population: about 1.4 billion - He notes China has entered population decline China projected population: around 700 million - His long-run estimate, implying roughly a halving Japan workforce decline: about 10 million fewer workers than in 1995 - Used to explain weak macro growth despite respectable per-adult performance Time lag for education reform: about 25 years - A reform today would only affect the labor force when new graduates enter employment
Pivotal Quotes: "if we did 10, had we done seven, we will probably be on a better place" — Jesus Fernandez-Villaverde: His view that COVID-era stimulus was directionally appropriate but quantitatively excessive "we are not reproducing ourselves" — Jesus Fernandez-Villaverde: His core claim about humanity having fallen below replacement fertility "the problem of central planning is not that given the information available, you find the optimal solution. The problem is how you get that information" — Jesus Fernandez-Villaverde: His explanation of the knowledge problem and why AI does not solve it
Implications: Listeners should expect lower trend growth, sharper regional inequality, political realignment, and more pressure on institutions as populations age and shrink. AI may boost efficiency, but it is unlikely to fully offset demographic decline or eliminate the need for markets and price signals.
About Macro Musings
Hosted by David Beckworth of the Mercatus Center, Macro Musings pulls back the curtain on the important macroeconomic issues of the past, present, and future.