Nicholas Bloom on Economic Impacts of COVID-19 in the Short-run and Long-run

Episode Summary

Executive Summary: Nicholas Bloom argues COVID-19 will cause an unusually severe U.S. recession driven as much by uncertainty as by the direct shutdown shock. He expects a sharp GDP collapse in 2020, a slow recovery, and lasting structural effects: higher risk aversion, lower productivity growth, reduced globalization, more inequality, and a possible shift away from dense urban living.

Main Topics: Measuring COVID-era uncertainty (Priority: 5/5): Bloom explains why traditional recession indicators are too slow and why real-time measures—stock market volatility, newspaper-based uncertainty, and surveys—are better suited to tracking a fast-moving pandemic shock. Short-run GDP collapse and weak rebound (Priority: 5/5): Using a VAR framework calibrated on cross-country historical data, Bloom predicts a severe contraction in U.S. output in 2020 and rejects a quick V-shaped recovery, expecting instead a U-shaped or slower rebound. Uncertainty as a major transmission channel (Priority: 5/5): He argues uncertainty is roughly as important as the direct first-moment shock, depressing investment, hiring, and consumer spending because firms and households delay decisions when the outlook is unclear. Long-run productivity and innovation effects (Priority: 4/5): Bloom warns that shutdowns of labs, R&D, training, and intangible investment will likely reduce trend productivity growth, especially because productivity gains have already been slowing for decades. Distributional and political consequences (Priority: 4/5): The crisis disproportionately harms lower-income, hourly, and self-employed workers, while salaried workers can often work remotely, potentially increasing inequality and shifting political pressures toward higher taxes on wealthier households. Globalization, immigration, and fiscal legacy (Priority: 4/5): He expects COVID to slow trade and immigration, both key drivers of U.S. growth, while also leaving the U.S. with much higher public debt that may eventually require higher taxes or inflation. Urban living and working from home (Priority: 3/5): Bloom suggests the pandemic could reverse some of the long-running premium on dense city centers as more people work remotely part-time and re-evaluate urban density and commuting patterns.

Key Arguments: COVID created an unusually fast macroeconomic break, so forecast models based on pre-2020 trends are poorly suited to the moment. Real-time uncertainty indicators (VIX, news-based uncertainty, surveys) are essential because GDP data arrive too late to guide policy or forecasting. The economic hit is not just supply shutdowns; uncertainty is a major driver because firms cut hiring, R&D, and investment while households save and avoid durables. Bloom’s historical model implies a GDP trough around -11% to -12% later in 2020, with downside risk potentially approaching Great Depression-like magnitudes within a year. The recovery will likely be slow because many activities—labs, training, startup formation, intangible investment—cannot restart instantly once shut down. Working from home can raise productivity in selected jobs, but mass remote work during COVID is different because of childcare, poor home workspaces, reduced choice, and limited job feasibility. The crisis may permanently raise risk aversion, which would lower entrepreneurial activity and support higher expected equity premia relative to interest rates. Trend productivity growth may slow further because U.S. innovation was already weakening before the pandemic and many R&D processes are now interrupted. The shock is regressive: hourly workers, contractors, and lower-income households bear disproportionate losses, while high-income salaried workers are better protected. Globalization and immigration likely weaken after the pandemic, removing two structural contributors to U.S. growth and innovation. The federal debt burden will rise materially, making future fiscal adjustment likely through higher taxes rather than inflation. Dense urban areas may lose some of their premium if remote work persists and people become more cautious about high-density living.

Data Points: U.S. unemployment rate (Feb 2020): 3.4% - Bloom cites February 2020 as the strongest labor market in roughly 60 years. Projected GDP contraction (2020 trough): -11% to -12% - Bloom’s VAR-based forecast places the U.S. economy at a deep trough later in 2020. Consensus forecast for Q2 2020 GDP growth: -4% quarter-over-quarter - He contrasts his forecast with consensus estimates for the second quarter. Consensus annualized Q2 2020 GDP growth: -16% - Bloom explains that quarterly growth of -4% annualizes to roughly -16%. Goldman Sachs Q2 2020 forecast: -6% quarter-over-quarter / -24% annualized - He cites Goldman Sachs as even more pessimistic than consensus on the annualized basis. VIX increase: about 400% above baseline - Bloom uses stock market volatility as a real-time measure of uncertainty during COVID. VIX level: over 80 - He notes the VIX rose above 80 from an average near 15 before the crisis. Economic Policy Uncertainty Index increase: about 4x - Bloom says the index rose from around 100 to around 400. Forecasted sales drops in surveys: minus 15% to 20% in late March; around 40% in a small-business survey - He reports worsening firm expectations over March and later fintech survey results. Moody’s estimate of daily GDP lost: roughly 30% of GDP per day - He cites a bottom-up estimate of output lost due to shutdowns. Employment gains since July 2009: about 20 million to 22 million jobs - He says the entire employment expansion since the Great Recession could be erased within weeks. Share of jobs workable from home: around one-third of employment - Bloom cites research by Jonathan Dingel and Brent Neiman. Share of wage bill workable from home: roughly 50% - Higher-paid jobs are more likely to be compatible with remote work. Productivity gain from working at home (Ctrip experiment): 13% - Bloom describes his RCT in China showing large gains for selected remote workers. Quit rate reduction in Ctrip experiment: about 50% - He notes remote work reduced attrition substantially in the experiment. Business startup decline: 40% - He says new EIN applications, a proxy for startups, fell 40% year over year. Government debt burden: about $25 trillion - Bloom estimates total U.S. government debt across federal, state, and local levels. Debt-to-GDP ratio: around 100% rising toward 130% to 140% - He expects the pandemic to push debt well above World War II-era levels. Top marginal tax rate in the 1950s: 90% - Bloom uses this to illustrate how debt burdens have historically been managed. Jobs lost in the shutdown: roughly 22 million - He cites forecasts that all job gains since 2009 could be wiped out quickly. Relative ability to self-isolate: Top 10% far more than bottom 10% - He references cell-phone data showing income-based differences in isolation behavior. Share of Silicon Valley PhDs who are immigrants: more than half - Bloom uses this to underscore the importance of immigration for innovation. Historic productivity slowdown: declining since the 1950s - He references broader secular slowing in U.S. and European productivity growth.

Pivotal Quotes: "“I think the short run is extremely bleak.”" — Nicholas Bloom: Bloom summarizes his view of the near-term macroeconomic outlook for the U.S. economy. "“If ever there was a turning point, it was March 2020.”" — Nicholas Bloom: He argues conventional forecasting methods fail when the economy abruptly shifts from expansion to crisis. "“The innovation engine is almost completely stalled right now.”" — Nicholas Bloom: Bloom explains why he expects weaker long-run productivity growth and a slow recovery.

Implications: Listeners should expect a deep recession, a slow recovery, and lasting structural shifts in work, cities, taxes, and growth. Firms and policymakers need real-time data, flexible forecasting, and plans for higher inequality and weaker trend growth.