The study by Angulo estimated that only about 1 in 4 COVID cases were confirmed (
https://jamanetwork.com/journals/jam...rticle/2774584). If we assume that ratio is accurate today then, based on the 48M official cases, we could estimate that about 112M in the U.S. have been infected (about 34%).
According to the CDC, 42M people have received at least one dose (
https://covid.cdc.gov/covid-data-tracker/#vaccinations), which is about 13% of the U.S. population. Obviously, the percent "protected" is less because some of that 13% were only recently vaccinated and many have not received their second dose, and the WSJ opinion piece doesn't mention that. Then again, the new data cited above in this thread suggest that one dose provides good protection after a few weeks. The effective rate might be a bit higher if the people being vaccinated tend to be those most likely to transmit COVID; my understanding is that kids are less likely to get and transmit COVID (e.g.,
https://www.nationalgeographic.com/s...ad-coronavirus).
Of course, you can't just add these two numbers, because they include some of the same people. If you assume independence, and a U.S. population of 328M, then the about 43% of the population has some protection, and rising. It makes sense that COVID transmission is dropping fast. Elimination by April an back-to-normal seems overly optimistic, though.
Notes: a lot of assumptions in there, so the confidence intervals are pretty wide. Also, the U.S. is obviously not just one homogenous blob. Trusting the 43% estimate, you would presumably have regions with 70% infected where COVID is nearly stamped out, and regions with 10% infected where it can still be easily transmitted.