I saw a post today comparing COVID-19 cases in the United States to the European Union, but since I don’t really trust case counts I skipped by it. Still, it got me curious, so I decided to take a look at COVID-19 deaths in the US vs. the EU. What I got was two curves that were pretty similar except that the US was about 12 days behind Europe. So I lagged the US data by 12 days to get a good visual comparison. Here it is:

That’s sort of an interesting look, isn’t it? Both the US and the EU rose at almost exactly the same rate and reached almost the same peak. The difference comes after the peak: the EU kept most of its lockdowns in place for about six weeks post-peak and crushed their curve to below 1 death per million. The US, by contrast, began opening up within three weeks after its peak, and our decline has been much less rapid.

Even if we don’t have a resurgence of infections in the near future, our rate of decline has flattened considerably since the end of April. At the rate we’re going, we won’t hit Europe’s level until the end of July.

Here’s some pretty good news. Retail sales rebounded very nicely in May:

Retail sales declined by $79 billion between February and April, but recovered 81 percent of that loss in May. This is only a single month, and it’s not evidence of a “v-shaped” recovery—not yet, anyway—but it’s still good to see. With money from both the stimulus checks and the unemployment bonus finally reaching consumers, wallets are opening up.

How much of this rebound was due to e-commerce sales? Unfortunately, that’s tallied quarterly and the next report won’t be available until July. But I expect to see that it played a pretty big role.

Note that the chart above is for all retail sales except food services. The story there, where it’s something you do in person or not at all, is a lot less buoyant:

Restaurant sales plummeted by more than half between February and April, declining from $65 billion to $30 billion. In May they recovered only 26 percent of that loss. Even as restaurants slowly opened up, consumers remained wary of going out to eat.

This photo was taken only 90 days ago, but it seems like another era. No masks, no worries, no nothing. Just out with friends for a little late night snack at the local 24-hour IHOP. But the server is already disappearing like a ghost, a portent of things to come.

Everyone’s favorite publication, the CDC’s Morbidity and Mortality Weekly Report, has some new data on COVID-19 incidence in the United States. Here’s the basic breakdown by age:

This is new. It suggests that the incidence of COVID-19 is roughly the same all the way from age 20 to age 79. I wonder what this would look like if you removed nursing home deaths, which are surely a big part of the incidence among those over age 80.

The CDC also reported on the death rate among those with underlying conditions, primarily heart disease, diabetes, and chronic lung disease:

Among middle-aged patients, those with underlying conditions have a far higher mortality rate. The total incidence of COVID-19 is low among 30-40 year olds, for example, but those who get it are a whopping 150x more likely to die if they have an underlying condition. For all age groups combined, the death rate for those with underlying conditions is 12x higher than those without underlying conditions.

And there’s yet another confirmation that COVID-19 has hit people of color far more heavily than white people:

The incidence of COVID-19 among non-white populations is nearly 3x that of the white population.

POSTSCRIPT: I sure hope I have all this right. The CDC reports this data in a cumbersome way and I had to convert from hospitalizations to the whole population. I’m pretty sure I did it all correctly, but if anyone notices an error, please let me know.

The Evil Dex.Kevin Drum

Hell, all they had to do was ask me:

Scientists at the University of Oxford said on Tuesday that they have identified what they called the first drug proven to reduce coronavirus-related deaths, after a 6,000-patient trial of the drug in Britain showed that a low-cost steroid could reduce deaths significantly for hospitalized patients. The steroid, dexamethasone, reduced deaths by a third in patients receiving ventilation, and by a fifth in patients receiving only oxygen treatment, the scientists said. They found no benefit from the drug in patients who did not need respiratory support.

The Evil Dex! The four-dollar pill that does everything! But this is only fair. My immunocompromised status makes me more likely to be hospitalized from COVID-19, but my long-time status as a dex user will help me survive. Funny how that works.

In related news, my doctor has approved a reduction in my dex intake to 2 mg. This is such a tiny dose it’s surprising my body will even know it’s there. Then again, my body has gotten increasingly sensitive to it over the years, so maybe all I need to do is wave a pill in front of my nose and that would be enough. Future M-protein tests will tell us whether 2 mg works.

UPDATE: Since I spend a lot of time griping about the side effects of dex, I should mention that they only appear after several weeks on the stuff. My guess is that patients with COVID-19 who are given dex for a week or two won’t suffer much in the way of side effects at all.

UPDATE 2: Via Twitter, Craig Kaplan writes: “Important to be very clear about Dex use as something that can only treat haywire immune response to coronavirus infection but likely would increase chance of infection if taken early. Timing will be everything and only should be given later in disease course to mitigate.”

In other words, don’t even think about taking dex if you aren’t in the hospital, hoping that it might protect you from COVID-19. It won’t, and it might even make things worse. It only works among hospitalized patients who are on oxygen treatment.

Here’s the coronavirus death toll through June 15. Mexico is spiking back up a bit, while the United States continues its slow crawl to 2 deaths per million. We’re roughly following the path of Sweden and the UK, lagged by a week or two. That’s not great, but then again, I never expected us to decline this far regardless of how long it took. So this is good news overall, but as I’m sure you all know, there are a whole bunch of states that have high and rising death counts following the reopening. It may be only a matter of time before those states swamp the others and push our overall mortality rate higher.

OPERATIONAL NOTE: When I first started doing these charts, the folks at Johns Hopkins published their final daily data at around 6 pm Pacific time. That was great. I’d update my charts before dinner and schedule them to appear the next morning. However, a while back it became 8 pm. Then 9 pm. Yesterday it was 10 pm. If this keeps up, I’m going to give up doing them the night before and instead update them first thing in the morning. Since I’m a pretty variable riser these days, this means they’ll appear at variable (but later) times than they do now. I’ll keep you posted if this happens.

The raw data from Johns Hopkins is here. The Public Health Agency of Sweden is here.

I have tried to stop posting about Donald Trump’s idiocy of the day, whatever it might be. I’ve really, really tried. But every time I try to get out, he pulls me back in. The topic today is the People’s Republic of Seattle:

REPORTER: What can you do?

TRUMP: About 10 different things, any one of which will solve the problem quickly.

REPORTER: Could we go through a whole list or a partial list?

TRUMP: We don’t have to go through any list…Before we do something, I’d like to see the press…cover it. pic.twitter.com/IG4ifIMxSr

— JM Rieger (@RiegerReport) June 15, 2020

The Navajo Nation really, really, really, really doesn’t want you to park on this bridge.

One of the reasons that recovery was so slow from the Great Recession of 2007-09 is that there were huge headwinds due to reduced government spending. You can read the whole story here, but this chart tells you most of what you need to know:

After most recessions, government spending goes up substantially, which helps economic growth. Republicans declined to allow that to happen while Obama was president, but they’re singing a different tune now that one of their own is in the Oval Office:

White House trade adviser Peter Navarro signaled that President Donald Trump is looking for at least $2 trillion in the next relief package being considered to help buoy an economy devastated by the coronavirus pandemic….Navarro did not say whether the bill should include additional relief for American cities.

Federal government spending to fight the coronavirus recession has been fairly healthy, but what about state and local governments? Michael Hiltzik reminds us that celebration was premature when the Labor Department announced an increase in jobs last week. Partly this is because it was a blip compared to the huge number of jobs that have been lost:

But the most important reason not to celebrate was hidden in the government report in plain sight. Employment by state and local governments has fallen off a cliff. The employment report issued June 5 by the Bureau of Labor Statistics showed that state and local government employment fell by 571,000 jobs in May. The month before, the loss was 964,000, for a two-month total of 1.535 million public sector jobs lost.

And the disaster may just be starting. Estimates of the size of the deficits faced by state and local governments through 2022 from the combination of heightened public health spending to combat the coronavirus and sinking revenues due to the economic shutdown and its continuing reverberations range from a catastrophic $500 billion through fiscal 2022 to a cataclysmic $959 billion through the end of next year.

It’s a nasty looking chart:

This is why it’s so important that the next round of spending includes assistance for state and local governments. Navarro mentioned that he wanted a payroll tax cut along with aid for manufacturing and pharmaceutical companies. But he mentioned nothing about state and local government. At this point, though, that’s probably the single most important part of any stimulus package. Democrats should be absolutely unwilling to even consider any legislation that doesn’t address this.

Hmmm:

The Supreme Court on Monday refused to hear the Trump administration’s challenge to a California “sanctuary” law, leaving intact rules that prohibit law enforcement officials from aiding federal agents in taking custody of immigrants as they are released from jail….The Trump administration’s challenge was launched by former Atty. Gen. Jeff Sessions. He insisted California was unconstitutionally interfering with federal immigration enforcement. But the Supreme Court, in a decision written by the late Justice Antonin Scalia, has said state and local officials are not obliged to carry out federal enforcement. That state’s rights doctrine appears to have prevailed. Even Trump’s two appointees — Justice Neil M. Gorsuch and Brett M. Kavanaugh — refused to hear the administration’s appeal.

That’s more good news from the Supreme Court today. It’s making me nervous. When will the other shoe drop?