The Washington Post has today’s least surprising headline:

This is so appalling. It doesn’t even matter what the reason is. Maybe it’s mass delusion. Maybe it’s cowardice. Maybe it’s just the usual cynical con designed to keep the conservative base outraged. Whatever the reason, the result is ever more senseless destruction of a political system that, for all its faults, is still fundamentally sound if all sides act even remotely reasonably. But for how much longer?

UPDATE: Now we’re up to 27!

Here in California, new lockdowns and travel restrictions are the order of the day. This extreme response to rising COVID-19 cases is based on ICU capacity, so it’s perfectly reasonable. However, it’s prompted a spate of articles about how the coronavirus is “raging out of control” in California with an astonishing 22,000 new cases recorded yesterday.

I don’t understand. This is roughly 10 percent of all new cases in the country, and California has a little more than 10 percent of the US population. It’s basically what you’d expect. To test this, I made a chart (of course I did):

This shows two things. First, as goes the country, so goes California. We’re just following the overall trend. Second, for the past couple of months we’ve been substantially below the national average. I suppose you can say that cases in California are “raging out of control” simply because COVID-19 is raging out of control practically everywhere, but that’s pretty deceptive. In fact, California is doing relatively well. As of yesterday, we were about one-quarter below the national average.

So what’s with the news coverage? Can’t these folks do simple division?

The US surpassed 6 deaths per million on Friday. Our trajectory looks scary as hell, though it’s possible that some of this is just making up for the artificial Thanksgiving dip. It might flatten out a bit by next week, though it will certainly keep climbing.

Here’s the coronavirus death toll through December 4. The raw data from Johns Hopkins is here.

Isn’t this adorable? Needless to say, the licking only lasted a few moments, at which point Hopper sensed a change in mood and gave Hilbert a little swat. Then they passed each other on the fence and went their separate ways.

Here’s an interesting chart:

This chart comes from a new paper by Taeho Kim of the University of Chicago, who investigated the rate of police-related homicides in cities that adopted body cams vs. those that didn’t. There are, of course, lots of things going on that could mess up these results, so Kim used an experimental design that depended on “idiosyncratic variation in adoption timing that is attributable to administrative hurdles.” This provides a fairly random way of creating a treatment group vs. a control group.

As you can see, agencies that adopted body cams saw a substantial reduction in police homicides: from 0.39 per 100,000 in 2013 to 0.27 in 2019. By contrast, agencies that didn’t adopt body cams showed no change.

And there’s more: virtually all of the decline in police homicides was due to reductions in killings of Black and Hispanic suspects. Nor was anything else affected. Arrests stayed at their previous levels, suggesting that body cams didn’t reduce policing effort. Kim also found that public opinion toward the police improved in jurisdictions that adopted body cams. On a purely monetary basis, Kim estimates that adoption of body cams is a big cost-benefit winner simply by reducing payouts from lawsuits over excessive use of force.

This is a very tricky bit of research, since there are lots of reasons that police homicides might go down in some areas and not in others. Nonetheless, by comparing lots of different agencies instead of changes in a single agency, Kim has advanced what we know about body cams.

The American economy gained 245,000 jobs last month and the unemployment rate declined slightly to 6.7 percent:

Overall, we’ve lost about 10 million jobs in 2020. As always, though, I’d caution everyone not to read too much into these numbers. November’s job gains are disappointingly small, but there’s really no reason to think we’re going to get the bulk of the lost jobs back until the middle of next year, when (hopefully) the pandemic recedes.

Nearly every industry showed (small) gains from October with one exception: government. We should expect that to continue unless Congress passes a relief bill that includes money for states and cities.

I guess I’m just perpetually confused these days. The Wall Street Journal—following the lead of lots of other news outlets—tells us today that labor force participation has dropped during the pandemic, and that it’s especially dropped for women. On the first claim, it’s hard to say anything other than “duh.” We have deliberately closed businesses and put people out of work, so of course labor participation is down. As for the second, here’s the chart:

The participation rate for men has dropped 1.6 percentage points. The participation rate for women has dropped 2.0 percentage points. This is not a big difference, and in any case it’s all due to noise between August and October. It could turn around completely next month. Or it could get worse. Who knows?

Journalists seem bound and determined to slice and dice economic data these days looking for tiny differences that they can spin into a narrative of some kind. They should stop. The fact that employment is down is inevitable, and it tells us nothing about whether it will recover when the pandemic recedes. And when you start slicing small numbers into ever smaller ones, it tells you less than nothing.

If there’s something big happening, then by all means tell the story. But if you’re relying on fractions of a percent in a noisy data series, it’s time to step back and think about whether you’re really sure anything serious is going on.

It looks like France has fixed its reporting problem. Germany has shot way beyond its mortality level from the first wave and looks set to continue rising. Here in the US our trajectory appears to be nearly straight up.

Here’s the coronavirus death toll through December 3. The raw data from Johns Hopkins is here.

Yeah, he looks tired now. But a few days of rest at the end of January should perk him right back up.Chris Kleponis - Pool Via Cnp/CNP via ZUMA

For a long time I’ve been basically optimistic about life after Trump. It’s certainly true that Trump is a symptom of the rot in the Republican Party, not a cause of it, but it’s also true that he has personally spearheaded a phase change, if you will, that has taken the party to a whole new level of toxicity. But he could only do that because he’s a unique personality. I figured that once he was gone the party would forget he ever existed and quietly settle back into its usual role of kowtowing to millionaires and white evangelical Christians. Trump, I thought, was a blip, not a harbinger of things to come.

I still believe that, but only if Trump actually goes away. That’s why stories like this unnerve me:

President Trump has raised $495 million since mid-October, with $207.5 million of it pouring in after Election Day — an extraordinary haul resulting from Trump’s post-election fundraising effort using a blizzard of misleading appeals about the integrity of the vote.

….Much of the money raised since the election probably will go into Save America, a political action committee that the president can use for various activities after he leaves office. Some of the contributions will go toward what is left of the president’s legal fights over the certification of election results, which have failed to gain traction in the courts.

If this really is pure grift, and Trump uses the money to build a new wing at Mar-a-Lago, I’ll breathe a sigh of relief. But probably not. This kind of fundraising suggests pretty strongly that Trump plans to stay very, very active—maybe on TV, maybe doing rallies, maybe creating a whole new cable channel just for himself and his buddies. And I don’t know how to think about that. If he does this, will he retain his deathlike grip on the Republican Party? Or will he eventually fade away, like Sarah Palin?

I don’t know. But I can say that I’m a little less optimistic than I used to be.

Today the New York Times points to a recent study suggesting that lots of people are qualified for jobs which would pay them way more than they’re making now. That’s probably true, and I don’t have any beef with it. However, the study’s abstract starts out like this:

The demand for a skilled workforce is increasing even faster than the supply of workers with college degrees – the result: rising wage inequality by education levels, and firms facing a skills gap.

This has been on my mind for a while, but I’ve never bothered writing about it before now. The thing is, there’s no question that college grads make a lot more money than high school grads. But whenever I look at income data, that gap isn’t getting any bigger—at least not at the level of high school and bachelors degrees. (PhDs are a different case.) Here’s a chart showing income growth since the end of the Great Recession:

Over the past decade, the income of women with BAs has gone up 9.6 percent, while women with high school diplomas have increased 8.2 percent. Among men, BAs increased 8.2 percent while high school grads increased 6.5 percent. (This is all adjusted for inflation, of course.)

The income of college grads has increased more than the income of high school grads, but only by the tiniest amount. I’ve zoomed in the chart so you can see the gaps, but at any normal scale there’s virtually no difference. The gap has widened slightly more than one percentage point since the end of the recession. If you go back two decades, it looks much the same.

And yet we keep saying that the educational income gap is rising out of control. Why? As near as I can tell, it’s because this was true in the 1990s and no one has bothered to adjust what they say about it since then. What am I missing here?