The Short Answer: It Depends What You Count
"How much did COVID cost the US?" has three defensible answers, because "cost" can mean three different things. They are routinely quoted interchangeably, and they should not be.
| If "cost" means | Answer | What kind of number |
|---|---|---|
| Economic output that was never produced, 2020–2023 | ~$14 trillion | A model estimate against a no-pandemic counterfactual (USC Schaeffer Center, 2023). Excludes deaths and illness. |
| Lost output plus deaths, long-term impairment and mental health, valued in dollars | ~$16.1 trillion | An October 2020 projection by Cutler and Summers in JAMA, about half of it health losses priced at $7 million per statistical life. |
| Federal money Congress made available for the pandemic, and what was actually paid out | $4.68 trillion made available; $4.57 trillion spent | An accounting figure: budgetary resources and outlays under the six COVID-19 relief laws of 2020–2021, as tracked by USAspending.gov. |
The first two measure what the country lost. The third measures what the federal government spent responding to the loss. That spending is not part of the $14 trillion and should not be added to it: relief payments moved money between people rather than destroying output. All three figures come from the sources in the table, read October 15, 2026, and the rest of this page works through the $14 trillion. How it reconciles with Cutler and Summers' $16.1 trillion is set out line by line on our hospitalization cost page.
The Headline Number
U.S. real GDP contracted 3.5% in 2020 — the steepest single-year decline since 1946, per the U.S. Bureau of Economic Analysis. That contraction captures only the first year of disruption.
A USC Schaeffer Center analysis by Hlávka, Rose et al. projected cumulative U.S. GDP running roughly 14% below its no-pandemic counterfactual through end of 2023 — a total economic cost of approximately $14 trillion. Source: Hlávka JP, Rose A et al. COVID-19's Total Cost to the U.S. Economy Will Reach $14 Trillion by End of 2023. USC Schaeffer Center. May 2023.
This estimate reflects a key insight: the direct health-care cost of COVID-19 — hospitalizations, treatments, vaccines — is a small fraction of the total. The far larger loss comes from behavioral change, supply-chain disruption, and the withdrawal of workers from the labor force.
What kind of number "$14 trillion" is — and what it is not
This distinction decides how much weight the figure can carry, and it is almost always omitted when the number is quoted.
The −3.5% GDP contraction is a measurement. It is BEA's advance estimate for calendar 2020, published January 28, 2021 — a national statistic produced from reported economic activity. BEA revises annual GDP figures in subsequent estimates and comprehensive updates, so the currently published value for 2020 may differ from the advance estimate quoted here; we cite the advance estimate and say so rather than quoting a revised figure we have not verified against BEA's current release.
The USC Schaeffer Center’s $14 trillion is a model output. It is the difference between actual cumulative US GDP and a counterfactual — what GDP would have been had the pandemic not occurred. No one observed that counterfactual. The estimate comes from a computable general-equilibrium model run by economists at the USC Schaeffer Center, and its answer depends on assumptions the modellers made about how the economy would have grown, how much of the behavioural change was pandemic-caused, and how long the effects persist. A different reasonable set of assumptions produces a materially different total. That does not make it a bad number; it makes it a projection with a stated method, not a statistic like the GDP figure above.
It is also not a government estimate. USC's Schaeffer Center is a university research centre, not a statistical agency. We cite it because it is the most widely used whole-economy estimate and because its method is published — but this page names it as academic modelling every time it appears, rather than presenting it as an official US figure.
Showing the arithmetic
Two comparisons make the scale legible. Both are our calculations from the sources cited on this page:
- USC Schaeffer’s $14 trillion ÷ 331.4 million US residents (2020 Census) ≈ $42,200 per person — roughly $169,000 for a household of four, spread across four years and borne unevenly.
- USC Schaeffer’s $14 trillion ÷ $6.6 trillion in total FY2020 federal outlays (CBO) ≈ 2.1 — the estimated economic cost is about two-and-a-tenth times everything the federal government spent in the pandemic's first fiscal year, on everything.
Neither figure is in the USC Schaeffer, CBO or Census sources; both are simple division, shown so the reader can check them. Neither should be read as a bill anyone received: most of the $14 trillion is output that was never produced, not money that changed hands.
Key figures at a glance
- U.S. GDP contraction, 2020: −3.5% (BEA, largest since 1946)
- Total U.S. economic cost through 2023: ~$14 trillion (~14% cumulative GDP below no-pandemic path) (Hlávka, Rose et al., USC Schaeffer Center, 2023)
Direct Health-System Costs
Direct COVID-19 medical costs — hospitalizations, ICU care, outpatient visits, and the national vaccine program — were historically large in absolute terms, yet modest relative to the total economic shock. Federal emergency health spending included hundreds of billions for hospital relief, testing infrastructure, and vaccine procurement and distribution.
The U.S. health system absorbed a surge in demand unlike anything since the 1918 influenza pandemic: at the peak of the Delta and Omicron waves, hospital capacity in many regions was strained to the point of delaying elective procedures and diverting ambulances, adding indirect costs beyond direct COVID treatment.
What one stay actually cost the hospital — $11,275 on average across 1.33 million stays, and how little insurance status moved it — and how the $14 trillion figure reconciles with Cutler and Summers' $16 trillion, is worked through on our hospitalization cost page.
Broader Economic Losses: Output, Labor, and Supply Chains
The bulk of COVID-19's economic toll was indirect — driven by three compounding forces:
1. Behavioral aversion
Even before government restrictions took effect, consumer spending on in-person services — restaurants, travel, entertainment, retail — collapsed as individuals voluntarily reduced exposure risk. This "aversion behavior" (spending and activity lost due to fear rather than policy) accounts for a large share of 2020's GDP decline and persisted in modified form well into 2021–2022.
2. Labor-force withdrawal
COVID-19 removed millions of workers from the U.S. labor force simultaneously — through acute illness, death, caregiving obligations, and early retirement. Labor-force participation fell sharply in spring 2020 and had not fully recovered to pre-pandemic trends by 2023, representing a lasting reduction in productive capacity. Long COVID has since emerged as a structural labor-force issue: the CDC estimates that a meaningful share of working-age adults who contracted COVID-19 experienced persistent symptoms affecting their ability to work.
3. Supply-chain disruption
Global production and shipping networks, optimized over decades for just-in-time delivery with minimal inventory buffers, proved acutely vulnerable to the simultaneous shutdown of factories, ports, and logistics hubs worldwide. The resulting shortages — semiconductors, consumer goods, building materials — cascaded through the economy for two to three years, contributing to the highest inflation since the 1980s and forcing costly restructuring of supply chains away from single-source dependencies.
Long-Term Economic Effects: Long COVID and the Workforce
Long COVID — post-acute sequelae of SARS-CoV-2 (PASC) — represents an ongoing economic cost that extends beyond the acute pandemic phase. Symptoms including fatigue, cognitive impairment ("brain fog"), and cardiovascular complications have affected workers' capacity to return to full employment, particularly in physically or cognitively demanding roles.
The economic cost of Long COVID's workforce impact is actively studied but not yet fully quantified. What is clear from NIH RECOVER Initiative research is that Long COVID is not a marginal phenomenon: it has affected millions of Americans and represents a long-term drag on labor productivity that does not appear in 2020–2021 GDP figures.
Putting the Cost in Perspective
Total U.S. federal outlays for fiscal year 2020 were approximately $6.6 trillion, per the Congressional Budget Office (Monthly Budget Review: Summary for Fiscal Year 2020, November 9, 2020) — outlays up 47% on the previous year. The same report puts the FY2020 deficit at $3.1 trillion, or 14.9% of GDP, the largest share since 1945, with federal debt held by the public rising to 100.1% of GDP. Against that, the estimated $14 trillion economic cost is roughly twice everything the federal government spent in the pandemic’s first fiscal year.
This scale explains why pandemic preparedness has moved from a niche public-health concern to a mainstream economic policy debate. The return on investment from early outbreak detection, vaccine stockpiling, and health-system surge capacity is measured not in lives alone but in trillions of dollars of economic output preserved.
For a cross-disease comparison — how COVID-19's economic cost compares to Ebola, norovirus, and hantavirus — see the Pandemic Economic Impact overview on VirusQuestions.com. For how the disease itself compares with 14 other viruses on exposure, hospitalisation and death, see our COVID-19 Risk Perspective and the hub's Virus Risk Perspective.
What Is Not Known
Economic questions about COVID-19 have a specific failure mode: the numbers are large, quotable, and mostly modelled, so a page can sound authoritative while resting on assumptions it never states. These are the gaps on this page, stated plainly.
- The counterfactual itself. Every "cost of COVID" figure is the gap between what happened and a modelled version of what would have happened. Nobody can observe the second half of that subtraction, and estimates published by different groups using different models are not directly comparable — which is why this page presents one estimate with its provenance rather than an average of several.
- How to separate pandemic effects from policy effects. Lockdowns, business closures, several trillion dollars of fiscal support (CBO), and the subsequent monetary tightening all occurred together. The USC Schaeffer $14 trillion figure is the net of all of it, and no published estimate cleanly attributes shares to the virus, to the public-health response, and to the macroeconomic response.
- The cost of Long COVID's labour-force effect. This is actively studied and genuinely unquantified. The national instrument that measured Long COVID prevalence — the CDC/NCHS Household Pulse Survey — published its final round for the period ending September 16, 2024 and has not been updated since, so even the denominator for such an estimate is now two years old.
- Distribution. A per-person average of about $42,200 conceals almost everything that matters. Losses fell disproportionately on in-person service workers, small businesses without cash reserves, and households that lost an earner. No single aggregate figure describes that, and this page does not pretend otherwise.
Where a number on this page is our arithmetic rather than a source's, it is labelled as such. When we get something wrong, the corrections log records it — including a correction to this page's BEA citation, made in August 2026.
Sources & References
- U.S. Bureau of Economic Analysis (BEA). Gross Domestic Product, Fourth Quarter and Year 2020 (Advance Estimate). January 2021. bea.gov
- Hlávka JP, Rose A, Walmsley T et al. COVID-19's Total Cost to the U.S. Economy Will Reach $14 Trillion by End of 2023. USC Schaeffer Center. May 2023. schaeffer.usc.edu
- CDC. Long COVID or Post-COVID Conditions. cdc.gov
- NIH RECOVER Initiative. Researching COVID to Enhance Recovery. nih.gov
- Congressional Budget Office. Monthly Budget Review: Summary for Fiscal Year 2020. November 9, 2020 — total outlays up 47% year over year; FY2020 deficit of $3.1 trillion, equal to 14.9% of GDP, the largest share since 1945; federal debt held by the public rose to 100.1% of GDP. cbo.gov
- U.S. Census Bureau. 2020 Census apportionment results — resident population of 331,449,281 as of April 1, 2020, used as the denominator in the per-person calculation above. census.gov
