Alarms and Excursions

Via NYTimes.

The Penn Wharton Budget Model at the University of Pennsylvania issued a new tool for estimating the costs and benefits of "reopening" states from their different lockdowns against the Covid-19 on Friday—costs being in infections and deaths, benefits in "economic growth" (scare quotes a function of having listened to this, which I hope to come back to) and regaining of lost jobs. Reporting on the thing is kind of nuts, as exemplified by this from Yahoo Business—

According to the Penn Wharton Budget Model (PWBM), reopening states will result in an additional 233,000 deaths from the virus — even if states don’t reopen at all and with social distancing rules in place. This means that if the states were to reopen, 350,000 people in total would die from coronavirus by the end of June, the study found....

However, partially reopening would also cause the death toll to rise, the university’s data found. An additional 45,000 lives would be lost, according to Wharton’s Budget Model, bringing the U.S.’s death toll from COVID-19 to 222,000.

—with the result that it could cause some needless hysteria on our side, and some embarrassment when it doesn't pan out, because there is no way all 50 states will "reopen" at the same time (reopening defined in terms of two factors, government relaxation of the rules and the public giving up its practice of social distancing), and the ones that certainly won't, in the Northeast and on then Great Lakes and Pacific Coast in particular, are the big population centers where the majority of those deaths would be happening. This shouldn't be regarded as a prediction, but as a rough estimate of what would happen if Trump had his way, which he won't, luckily. Another caveat for the benefit side:

However, the policy of reopening states would provide a much needed economic boost, according to the model.

“Almost all net job losses between May 1 and June 30 would be eliminated,” the report found.

Those jobs weren't going to be permanently lost anyway, or not all of them. It's impossible to say with certainty what economic life in the US is going to look like a year from now (I've been steeling myself to try to talk about that some), but the most you can say for the effects of getting the lockdown over with early is that it will hurry up the job growth effect, not fundamentally change it.

What's really likely to happen on the PWBM model is something more like this:

• The baseline predictions, that we'll reach a minimum total of something over 115,000 human lives lost to Covid-19 by the end of June, and nationwide weekly job loss rate of close to a million jobs, are presumably correct
• Florida, expected to get to 2,151 deaths, will get to 7,104 instead, assuming a partial reopening  and the end of the population's willingness to stay home
• Georgia, expected to get to 2,932 deaths, to 9,247
• Louisiana, from 2,523 to 3,348
• Alabama, from 650 to 1,946
• Mississippi, from 773 to 1,848
• Texas, from 1,642 to 7,015
• Arizona, from 772 to 3,818
• Iowa, from 951 to 1,748
• Kansas, from 678 to 2,925
• Missouri, from 597 to 1,932
• and comparable numbers for smaller, more remote and less urban states opening early

So judging on the basis of these possibly top ten states, the "reopening" we're currently experiencing  will kill something on the order of 30,000. Governor Kemp, Abbott, and DeSantis demanding the deaths of five or six thousand each (and Texas will be reaching positive weekly job gains by 8 May and Florida by 18 May in any case). And total national death toll totals of around 150,000.

Less, in fact, than the HHS/FEMA task force is predicting, which is a daily death rate increasing to around 3,000 by the end of this month on the basis of a model that's consistently underestimated death rates though it has performed very well on number of infections.

Via nytimes (but I think not behind paywall). 

But more than necessary, unless your aim was maybe to hurry up the process in view of a coming election or something like that, if you know what I mean.

I do wish the journalists would learn to read this stuff, though.