Take the timeline in order, because the order is the whole answer.
February 8, 2020, diary entry: he and a colleague are 'on the same page in thinking tht this is acting like a bad influenza in its transmissibility and that the denominator is much greater than 34,867 (above) making the case fatality rate (CFR) more like 0.2-0.3 % rather than 2.0%' [2].
February 28, 2020 - twenty days later - the New England Journal of Medicine publishes 'Covid-19 - Navigating the Uncharted,' by Fauci, Lane and Redfield. Under his own name, in the most widely read medical journal in the world: 'If one assumes that the number of asymptomatic or minimally symptomatic cases is several times as high as the number of reported cases, the case fatality rate may be considerably less than 1%.' And: consequences 'may ultimately be more akin to those of a severe seasonal influenza (which has a case fatality rate of approximately 0.1%) or a pandemic influenza' [1].
Same reasoning as the diary - an undercounted denominator - same downward direction, same seasonal-flu comparison.
March 11, 2020: House Oversight testimony. Reported mortality 'about three percent,' and counting asymptomatic infection, 'somewhere around one percent, which means it is 10 times more lethal than the seasonal flu' [3].
There is no private-versus-public split. The low estimate was published eleven days before the testimony. What moved between February and March was the number, from 'considerably less than 1%' to 'around 1%' - which is what an estimate does when data arrives.
Two smaller corrections. The '10 times the flu' line was said to Congress, not only on television, and it is arithmetic performed on the low number rather than a substitute for it: one percent divided by one-tenth of a percent is ten. The sentence producing 'ten times' is the same sentence revising mortality down from three percent. All three figures are also case fatality rates - nobody switched measures mid-argument.
One more thing belongs here, because a fact-check that only defends is not a fact-check.
The diary contains an entry that is genuinely difficult for Fauci, and the commentary has largely skipped past it. On March 15, 2020 he records speaking with New York's mayor and 'convinced him... to close the NYC schools,' adding that he told him 'he should close the bars and restaurants,' and describes a similar call with the chief of staff to California's governor [4].
If a person has publicly minimised their role in school closures, those entries are the problem - not the mortality estimate. This newsroom has not been able to source the exact wording or date of any such denial, so we are not asserting a contradiction. We are noting that the sharp claim being made is the one that fails, and the documented one is being under-reported.
How the estimate turned out is the last fact. More than 1.22 million Americans had died of COVID-19 as of March 6, 2025 [5]. The private figure was too low.
