This week we have another guest cat. This lovely gray critter is, unfortunately, an unwelcome visitor at my mother’s house. He’s obviously a well-fed cat, but he doesn’t seem to play well with others and all the other cats just hiss at him. When he runs away, he goes in a different direction each time, so we have no idea where he lives. It’s yet another mystery for these mysterious times we live in.

A few days ago I noted that low-income workers were still spending a lot, thanks largely to assistance from the CARES program, but they’re also still unemployed at high levels. My evidence was a little indirect, though, so I’m happy to present direct evidence for this based on data from Raj Chetty’s economic tracker. Here’s consumer spending:

The lowest income group has been spending at pre-pandemic levels since June, while the highest income group is still holding back. But here’s employment:

Low-wage workers are still far below their pre-pandemic levels of employment, while high-wage workers are back to normal and even mid-wage workers are close to normal.

Today, economists at Goldman Sachs became the latest to issue a warning that economic recovery will falter if Congress doesn’t provide more federal aid. That’s nearly a unanimous view these days, but these charts make it clear what kind of federal aid we need: not broad-based stimulus, but specific assistance to those who have lost their jobs. Let’s get with it.

Since late spring, police killings of unarmed people have started to trend down. Here’s the data from the Washington Post’s fatal force database:

Up through May, police shot and killed 28 unarmed people. In the four months following the death of George Floyd on May 25th, police have killed three. 

It’s a strain to draw any conclusions from such a tiny sample,¹ but I’m still struck by this. It’s an 87 percent drop in police killings of unarmed people, and it began right around the time that protests over Floyd’s death were spreading nationwide. As a comparison, police shootings of armed suspects haven’t changed much: They continued during the summer at the same level as before and continued to disproportionately impact people of color. This suggests that the drop in killings of unarmed people isn’t due to any larger trend in police becoming less lethal. It’s a separate phenomenon.

Very tentatively, then, we may now have a bit of evidence that policing has begun to evolve in response to America’s season of civil unrest over unjust killings. The big picture of overall police killings still hasn’t changed, but collectively police forces do seem to be showing greater restraint about shooting unarmed people—a legally and morally fraught problem, to say the least. If this keeps up, it suggests real impact from the Black Lives Matter protests, and is good news for everyone.

¹Although, for you stats nerds out there, the difference is significant at the p =.01 level.

Here’s the coronavirus death toll through September 24. The raw data from Johns Hopkins is here.

Europe does not look good. France is obviously showing an upward spike after more than two months at zero, and both Italy and the UK are spiking upwards slightly too. Worst of all though, is that if you squint hard enough you’ll even see a teensy little spike in Germany. Germany! The star pupil of COVID-19 containment. Hopefully these are all just blips, but it doesn’t seem likely that everyone would blip at once just by coincidence.

Here in the United States, we continue to cruise along at about two deaths per million. The trendline bounces up and down a bit, but never really seems to change much. Since August we’ve averaged a pretty steady 800-900 deaths per day from COVID-19.

The Washington Post says that Donald Trump is bullish on our recovery from COVID-19:

The president is betting his political future on convincing voters that a recent dip in coronavirus cases, hospitalizations and deaths signals a coming end to the national nightmare — and is a reason to reassess his handling of the pandemic and vote for him. He is zigzagging the country for a series of packed events, sometimes indoors, always with a packed and mostly unmasked crowd, preaching that the situation is rapidly improving while largely ignoring a death toll that this week surpassed 200,000 Americans amid fears that the country could have a second wave as temperatures drop.

I don’t get it. What dip in cases could he be thinking of?

New cases have been rising for the past two weeks, from 34,000 per day on September 13 to 43,000 today. And guess where things are worst?

The number of new Covid-19 infections in rural America jumped by 30% last week, reversing a short-lived decline in new cases and putting a record number of rural counties on the red-zone list. New deaths from Covid-19 also grew last week, increasing by 20% and bringing the total number of rural Americans who have died as a result of the pandemic to 18,128…The number of rural counties on the red-zone list climbed to 909 last week. The previous record was set the first week of September, when 806 rural counties were on the list.

These are Trump’s people. But who are they going to believe, Trump or their own eyes? Unfortunately, I have a feeling we all know the answer to that.

This picture was taken in the red rock country of northern Arizona, and it’s . . . a red rock. The photo required a 20-second exposure in order to capture the stars, and there was a single, dim light across the street from the rock that turned out to be perfect for such a long exposure. This is, needless to say, a bit of an austere picture, but I kind of like it for that reason.

What kind of economic growth can we expect if Joe Biden wins the presidency in November and Democrats sweep Congress? And how does it compare to a hypothetical Trump victory and a Republican sweep? The boffins at Moody’s Analytics took a look at the likely impact of economic policy in both scenarios and came up decidedly bearish on Republicans:

The economic outlook is weakest under the scenario in which Trump and the Republicans sweep Congress and fully adopt their economic agenda. In this scenario…it is not until the first half of 2024 that the economy returns to full employment…unemployment remains persistently higher…labor force participation rate that never fully recovers to its pre-pandemic highs…real after-tax income does not change much during the president’s term…The economy suffers in Trump’s second term, as we expect he will double down on the foreign trade and immigration policies he pursued in his first term…The significant increase in tariffs during Trump’s first term—from an effective tariff rate of 1.5% when he took office to a peak of more than 6% just prior to the Phase One deal—acted like a tax increase on the U.S. economy, hurting U.S. manufacturers, transportation companies, and farmers in particular. More of the same is expected in Trump’s second term.

Whew. That’s pretty brutal. Here’s a comparison in brightly colored chart form:

In every possible category, a Democratic sweep is better for the country than any other scenario. Moody’s even projects that Democrats would be better for the budget deficit than Republicans.

You can read the full report here, but it’s pretty easy to summarize. If Democrats win, they’ll spend money to stimulate the economy out of its COVID-19 funk and this will help everybody. The spending will largely be financed by taxing the rich, which has only a small negative effect on the economy. But if Republicans win, they’ll keep the purse strings closed and instead pursue yet more tax cuts for the rich and trade wars with China. Neither one is especially good for the economy. It’s so simple.

Brian Cahn/ZUMA

I was mulling over QAnon recently, wondering just how many people actually take the crackpot conspiracy theory seriously and approve of it. So I went looking for polls and found two, both conducted during the first half of September. They agree that only about half of Americans have even heard about QAnon, but they differ wildly in their assessment of how many fans it has:

• NBC News says that 3 percent of respondents are “very or somewhat positive” towards QAnon.
• Pew Research says that 9.4 percent of respondents think QAnon is “very or somewhat good for the country.”

That’s a big difference. If it’s really 9.4 percent, we have a serious problem. That’s a tenth of the population believing the country is under attack by a deep-state child sex trafficking ring. But if it’s 3 percent—well, that’s just business as usual. You can probably find 3 percent of Americans to support just about any wackadoodle conspiracy theory.

It seems like it might be important to get a firmer grip on this, especially since President Trump has recently praised QAnon and at least two Republican candidates for Congress have endorsed it, even as the FBI has identified it as a domestic terrorism threat. If QAnon is truly going mainstream, as Ali Breland wrote for us recently, it deserves some serious attention. On the other hand, if it’s still a fringe theory that’s just having its 15 minutes in the sun, maybe it’s not worth worrying about so much.

But which is it?

Section 230 of the Communications Decency Act of 1996 is foundational to the modern internet. What it says is that internet content providers aren’t responsible for content or comments posted by their users, even if the companies referee that content. This allowed companies like AOL and Facebook to step in and halt things like flame wars or racist diatribes, and to do so however they saw fit without having to worry that they could lose their status as mere common carriers that can’t be sued for things their customers say. Without this legal protection, no one would dare moderate anything and the internet would be a bottomless cesspool.¹

Lately, of course, this has become something of a political punching bag, largely because of conservative complaints that Facebook and other social media companies are unfairly targeting conservatives for censorship. This is fairly laughable if you know anything about the way that ultra-conservative nutbaggery dominates social media these days, as demonstrated daily by Kevin Roose’s roundup of Facebook’s top ten posts. But the Trump administration is nonetheless going forward with a proposal to amend Section 230:

The proposal advances two main goals the Trump administration and the department outlined in June: encouraging online platforms to actively address illicit conduct and manage content on their sites in fair and consistent ways. The department refined its proposal in the intervening months based on feedback from market participants and other stakeholders such as victims’ rights groups. As a result of that process, the department made some changes, including clarifying that internet companies would have immunity when they take down material that promotes violent extremism or self-harm, the official said. President Trump is also scheduled to discuss “protecting consumers from social media abuses” at a meeting Wednesday with state attorneys general, according to the White House.

….Mr. Trump and GOP lawmakers have complained about what they say are biased decisions to censor social-media posts or block certain users. Democrats, including presidential nominee and former Vice President Joe Biden, say platforms need to do more to curb the spread of false information.

You can safely ignore the stuff about illicit conduct. That’s pretty uncontroversial and bipartisan. The key addition is a clause that forces social media companies to regulate content “in fair and consistent ways.” The proposal spells this out as (a) having clear terms of service, (b) applying those terms of service to everyone, (c) not removing material on pretextual grounds, and (d) explaining all decisions clearly.

This proposal isn’t going anywhere until next year at the earliest, so there’s no special reason to examine it in detail until after the election. For all anyone knows, it might be little more than symbolic anyway, sort of a shot across the bow that will please conservatives even if it never becomes law. But it’s still something to keep an eye on. 

¹That is, even more of one than it is now. 

Here’s the coronavirus death toll through September 23. The raw data from Johns Hopkins is here.