Kevin Drum

Consider the following chart. It shows the value of commercial property under foreclosure:

What conclusion do you draw from the right edge of the chart, which shows foreclosures going up? Here's what the Wall Street Journal says:

Surge in Commercial-Property Foreclosures Suggests Bottom Is Near

In previous downturns, comparable surges in foreclosure activity has signaled the approach of a market bottom. Once lenders seize a property, they are typically quick to sell it, a process that helps determine values of properties after long periods of sluggishness in the sales market.

Does this make any sense at all? What goes up must come down, so foreclosures must be close to going down.

Remarkably, the author doesn't even try to provide evidence for this. And as usual, the Journal's original chart isn't adjusted for inflation, which makes the current surge look sharper than it really is. When you adjust for inflation it looks a lot less like it's reached a peak.

The longer I read the financial press the more I wonder what they're thinking. The actual evidence in this case suggests that foreclosures are going to increase for a while. The Journal itself reported a few months ago that delinquencies would rise at least through 2025 and probably longer. In fact, it was reported by the exact same person who wrote today's story. And office vacancy rates are currently higher than they were during the Great Recession, which suggests foreclosures probably have a fair ways to go.

In any case, no one knows for sure. But it beats me how someone can just declare that the bottom is near based on essentially nothing. Sometimes it's just vibes all the way down.

Here's a remarkable headline from NBC News:

These claims are repeated in the text, so it isn't just a matter of the headline being imprecise in order to save space. And all of them are wrong:

• Inflation has nothing to do with this. The cost of childcare has grown less than overall CPI and less than average wages.
  .
  
• It's not "American families." It's families with kids under 13 who use childcare. This is something on the order of 5-10% of all American families.
• The median family income in 2022 was $93,000. This makes childcare about 10% of the median income, not one-fifth.

How about the average monthly cost of childcare, which the Fed puts at $800. Has that gone up over the past few years? The Fed didn't track this in previous reports, but BEA data suggests probably not. There's no crisis and no chaos here.

In general, that is. NBC naturally found a family who says their spending on summer child care has more than doubled since last year. It's probably true! But it's also wildly unrepresentative of the overall situation.

We all know the old saying, "If it bleeds, it leads." News organizations have always focused on the sensational and the horrible, which is why we get so many stories like this one. Even in sober financial spaces, reporters insist on grasping for eyeballs with sensational claims that scare the hell out of people. The problem is that most of them aren't true. As with so many other things in America, childcare in general is about the same as it's always been.

Did I visit the Sahara Desert recently? No, just the Cadiz Dunes Wilderness near sunset, right here in Southern California. It's out in the middle of the Mojave Desert, sort of between Barstow and Needles.

I was there to see if it would be a good place for astrophotography and Milky Way pictures, which it turned out not to be. Partly this is because the dunes themselves aren't all that accessible and partly because the drive out there was considerably more of an adventure than I had counted on. More about that later.

Here's a chart showing how many of us have either gotten COVID or been vaccinated against it:

(Note that the line on the chart shows inverse seroprevalence. That is, the number of people who haven't had COVID and haven't been vaccinated. This allows me to draw a proper trendline, but you have to subtract the numbers on the trendline from one to get the actual seroprevalence.)

The trendline suggests we're now at 99.4% seroprevalence in the US population. We've finally reached the fabled level of herd immunity!

Sadly, it doesn't matter. It turns out that neither vaccines nor previous infections stop the spread of COVID. They just make it less dangerous. Oh well.

POSTSCRIPT: In case you're curious about how this breaks down, roughly 60% of the population has had COVID. (Almost) all the rest have been vaccinated.

Over the weekend, Jack Herrera had a piece in Politico that provides an interesting perspective on the Biden administration's immigration policy. It's true that illegal border crossings soared after Biden took office, but expulsions soared too, starting from his first day:

Most Americans don’t understand how many people the Biden-Harris administration has removed from the country.... In the spring of 2021, deep in the depths of the Covid-19 pandemic, I was in a camp in Tijuana, where some migrants were so hopeful the new president would let them in that they flew “BIDEN FOR PRESIDENT” flags outside their tents. But most of them who crossed got a slap from reality: They were quickly frog-marched by U.S. Border Patrol back through the deportation doorway, back to the squalid camps in cartel turf. Others got rapidly loaded onto ICE planes and flown back to Haiti, Guatemala, El Salvador, wherever. As the number of people crossing the border grew during Biden’s first two years in office, these expulsions reached a scorching pace. ICE charter flights bounced around the globe like Taylor Swift’s jet.

Naturally you want to see this in chart form. I'm here to help:

During Biden's term in office, he will have expelled, detained, or deported 6.9 million illegal immigrants. That's 73% more than Donald Trump.

This is why, despite the surge in migrants, the total population of illegal immigrants in the US has probably increased less than most people think. Out of roughly 11 million migrants (border crossers plus asylum seekers), around 7 million of them were expelled. At the same time, about 2 million illegal immigrants who are already in the country return to their homes every year. That gets us to 9 million. So net population growth has probably been on the order of 2 million or so, an increase of 18%.

That's just a guess, and it might be off on either side. Still, you might wonder why, if this is the case, Biden hasn't bragged more about it. I imagine there are two reasons. First, it's still growth of 2 million illegal immigrants compared to roughly zero during the Trump administration. Second, the Democratic base doesn't want to hear that Biden has been pretty strict about expelling illegal immigrants. It's a no-win problem.

The latest micro outrage on the right is about a performance at the Paris Olympics opening ceremonies. It involved some drag queens at a table full of people that supposedly resembled The Last Supper, and the New York Times says that organizers "have remained largely vague about the references associated with the images."

Then there's this brief reference seven paragraphs down:

It is unclear whether the drag queens scene was intended to reference “The Last Supper.” The official account of the Olympic Games said on the social media platform X that the scene with the blue-painted man, the French actor and singer Philippe Katerine, was an interpretation of Dionysus, the Greek god of wine and revelry, which “makes us aware of the absurdity of violence between human beings.”

Am I being obtuse here? If the official Twitter feed says this tableau was some kind of takeoff on Dionysus and classical myth, then there's nothing vague or unclear about it. And that's what it said in real time:

This was posted as the scene was televised. It's not some post facto effort to weasel out of responsibility. Thomas Jolly, the artistic designer of the show, provided this explanation on French TV:

[The Last Supper] was not my inspiration and that should be pretty obvious. There’s Dionysus arriving on a table. Why is he there? First and foremost because he is the god of celebration in Greek mythology and the tableau is called "Festivity," He is also the god of wine, which is also one of the jewels of France, and the father of Séquana, the goddess of the river Seine. The idea was to depict a big pagan celebration, linked to the gods of Olympus, and thus the Olympics.

This is pretty detailed. It's obviously not something you could make up on the spot. Given all this, how did the whole Last Supper thing gain so much traction?

POSTSCRIPT: I'm still trying to suss this out. Perhaps the tableau was meant to illustrate The Last Supper morphing into an episode of Dionysian revelry, and that's now being deliberately obscured by focusing only on the Dionysus scene? I suppose Jolly didn't explicitly rule this out, though it's a stretch. But maybe?

Here are a few assorted short takes that are on my mind right now. They are worth exactly what you paid for them.

Is Donald Trump losing his edge? Kamala Harris has been the Democratic nominee for a week now, and she's hardly immune from criticism. But so far Trump has called her a "bum" and "the most liberal person ever in US history"—not exactly biting attacks. What's going on?

I just finished a biography of John von Neumann, and it prompts a question: Is game theory actually useful for anything? It got famous in the '60s as the foundation of Cold War nuclear planning, but it never really told us anything we didn't already know without the math. I gather it's had some useful things to say about the design of auctions, but what else? There are some things like kin selection that you can explain in a game theoretic way, but you can usually explain them in other ways too. So what is its real-world value?

Why do so many people object to their content being used for AI training? I understand copyright infringement, if that's going on, but it generally isn't. The words and images are just used as part of an ocean of input that makes AI better. Why would I care, for example, if Google or OpenAI trawled my blog and turned it into tokens for use in AI training? Am I losing anything?

Am I the only one who didn't care much for the Paris Olympics opening ceremonies? My main complaint is that, until the very end, it was essentially a pure TV event. You couldn't really watch it in person since it was so spread out, and it included video snippets that only made sense (barely) as part of a TV show. That didn't sit right with me. (My other complaint is that so much of it seemed amateurish. How can that be with the kind of budget they have to work with?)

By the way, I'm now done with four weeks of radiation therapy (two to go) and so far I've had no big side effects to speak of. Hooray. I've been getting more and more tired, though, but I'm not sure if that's because of the radiation or the hormone therapy.

Today Donald Trump promised to create a strategic bitcoin reserve if he's elected. Even for Trump, does this make any sense?

Put aside the question of whether bitcoin is a real thing or just a mass delusion. Let's assume it's real. We still don't stockpile every real thing. We don't have a strategic Lego reserve or a strategic airplane reserve. We only stockpile things that we might suddenly need in large quantities.

We stockpile oil because we might need it if OPEC cuts off supplies. We stockpile vaccines because they take a long time to manufacture and we might need them quickly in case an epidemic strikes. We stockpile rare earth minerals in case China decides to get nasty about them.

But what conceivable emergency would prompt a sudden need for bitcoin? And if, for some reason, we did suddenly need a huge amount of bitcoin, what would prevent us from just buying it on the open market? The whole point of crypto is that it's decentralized and under the control of no government.

Before you tell me: Yes, I know this is just Trump pandering to his fans. Still, his actual policy pronouncements generally make some sense. This one is like promising to create a strategic baseball card reserve. Is there any possible way of interpreting it that isn't completely out to lunch?

What's up with the widespread mispronunciation of Kamala Harris's name?

The origin is simple enough. In English it's most common for the stress to be put on the second-to-last syllable of a word. This means a two-syllable word most often has the stress on its first syllable. A three-syllable word most often has it on the second syllable. So by default you'd pronounce Harris's first name as:

Kuh-MALL-uh

But this is far from a universal rule, and Kamala isn't an English name anyway. It has the stress on its first syllable:

KOM-muh-lah

It's a perfectly ordinary mistake for an English speaker to say this wrong. The real question is why so many conservatives who do know how it's pronounced are aggressively continuing to say it wrong. What's the point?

There isn't one, really, except to show that, by God, they aren't going to take orders from some liberal woman. Plus there's the usual Trumpism of refusing to ever admit a mistake, no matter how slight or obvious. And their deep habit—practically subconscious at this point—of disrespecting women and people of color. As a result, we have crowds of MAGA fans chanting kuh-MALL-uh as if they're showing her who's boss. It's kind of pathetic.

The Washington Post has a story today about the fact that medical providers are now required to make test results accessible to patients as soon as they're available (generally via phone or online portal). This is a very popular rule, but it can cause problems if the results are (or seem) scary and it takes days before you can talk to your doctor about them.

All true. But why does it take days for doctors to call and let you know what the tests mean? And why is it all but impossible to call a doctor yourself?

I'm here to tell you the answer. But first, a reminder that here in the US we spend more—way more—on health care than any other rich country in the world:

And what do we get for all this money? Not a lot of doctors:

This explains why doctors don't get back to us quickly: we don't have very many and they're overworked. We're nearly 30% below the average of rich countries. But it's actually even worse than that. Check this out:

Among primary care physicians, the US is dead last by a considerable margin. We are more than 70% lower than the average of rich countries. And believe it or not, it's even worse than that:

Not only do we have a minuscule number of primary care doctors, but the number has been essentially flat for more than 20 years while other countries have been steadily adding primary care doctors.

So what do we spend all our money on? The answer is:

• We pay doctors and nurses more than other countries.
• We pay specialists fantastically more than other countries.
• We pay more for pharmaceuticals than other countries.
• We pay more for machines than other countries.
• We pay a cut to insurance companies.

As a result of all this we can't afford to have a lot of doctors—and the ones we do have all want to become specialists because the pay isn't just better, it's massively better.

And that's why it takes a long time for your doctor to call you back.

POSTSCRIPT: In case you're wondering, even with the enormous salaries we pay specialists we're below average on per capita numbers even there. But only by about 4%, which means that specialists don't really seem to have a workload excuse for not calling back quickly.

Give or take a bit, primary care physicians in the US see 20 patients per day while specialists see 2.5 patients per day. Specialists spend a lot of their time doing procedures, of course, but with a patient workload like that you'd still think they could be a little more available. I'm not sure what their problem is.