Month: July 2024

A few minutes ago I ran across a chart showing that investment in factories had exploded recently and was twice as high as it had ever been in history. It was adjusted for inflation, which is great, but for something that big it struck me that it really ought to be shown as a percent of GDP. So I did:

By a funny coincidence, it turns out that in the second quarter of 2024 we did indeed set an all-time record: 0.854% of GDP compared to the previous peak of 0.847% in 1979.

Putting this into perspective, after falling during Donald Trump's term the rate of factory building has increased 2.5x since Joe Biden took office:

In dollar terms, we're currently spending $244 billion per year on factories compared to $91 billion three years ago. If you believe in restoring America's manufacturing capability, the past three years have been a golden era.

From Oklahoma's state superintendent of schools:

Huh. So declining to teach the Bible means you hate America. That's a fascinating point of view.

Naturally Walters says he wants the Bible taught solely for its "historical, literary and secular value." Nothing to do with proselytizing Christianity, no sirree. It's just a random historically important book.

The only surprising part of this is that Walters hasn't been sued yet, since that's obviously his intention. Maybe because the school year hasn't started yet, and no one can claim any harm until then? If so, we have three weeks to wait.

The New York Times/Siena poll has always been very Trump friendly, but that changed almost overnight after Joe Biden dropped out. Here is today's poll compared to the previous one:

Among likely voters, Kamala Harris made up five points in the national poll instantly. She's still behind, and swing states are still a problem, but that's still pretty spectacular progress.

At this point, I suspect the race hinges mostly on the 12-15% of people who say they're either unsure or planning to vote for RFK Jr—especially among Democrats. If, as I suspect, they mostly come home, Harris will end up with 50-52% of the popular vote.

Is that enough? I don't know.

The current hot topic in politics-land is who Kamala Harris will choose as her VP. And that's fine. It's good, clean fun.

But it's odd that the discourse is all so thoroughly conventional, focused almost entirely on demographic considerations. She needs a white man to balance the fact that she's a Black woman. She needs a moderate to balance the fact that she's a liberal. She needs a Midwesterner to balance the fact that she's from the coast. She needs someone from a swing state who will help capture crucial votes.

But help me out here, political scientists. Is there any evidence that VPs affect the vote even slightly? The last one who comes to mind is LBJ, who probably helped Kennedy win Texas in 1960. That was two-thirds of a century ago. Since then, though? Maybe John McCain's choice of the plainly unqualified Sarah Palin hurt him in an election he was going to lose anyway, but aside from that there's bupkis.

What nobody ever seems to mention is that Harris should pick someone who (a) she already knows and likes, and (b) would make a plausibly good president if she got run over by a bus. That's what I'd do. Since it doesn't really affect the election one way or the other, why not just choose someone whose advice she genuinely values?

A large RCT mask study was recently conducted in Norway. It involved about 4,000 people, half of whom wore surgical masks when they were outside the house and half of whom didn't. Here are the results:

The folks who wore masks got COVID at a slightly higher rate than those who didn't.

On the bright side, the mask wearers apparently got fewer colds and other respiratory diseases. So that's good.

But if it's COVID you're worried about, this is yet more evidence that it's N95 or nothing. Surgical masks and cloth masks have either small or no protective effects.

POSTSCRIPT: The study lasted two weeks, which suggests that over the course of a year more than a quarter of all people get COVID. That's a lot! Be careful out there.

This is Gustav Klimt's Judith at the Belvedere Museum in Vienna.

We now have all the important economic indicators for the second quarter of 2024:

• Real GDP grew 2.8%
• The unemployment rate was 4.0%.
• Inflation was down to 2.8%.
• Real wages were up 1%.
• Real disposable personal income was up 1.3%.
• Real consumption was up 1.5%.

What's not to like?

I promise not to turn this into a '70s music blog, but over the course of my various peregrinations this week I now have four or five hours of "70s on 7" under my belt. And on the drive to Anaheim¹ this morning it occurred to me that during this time I haven't heard a single Fleetwood Mac song. Also nothing from Bruce Springsteen. Nothing from Abba. Nothing from Pink Floyd. Nothing from Wings. Nothing from Steely Dan. Nothing from the Bee Gees. Nothing from Elton John.

I'm not saying these are all great bands. Opinions will vary. But they were all big acts in the '70s. I suppose it's praiseworthy that "70s on 7" isn't saturated with just the stuff we've all heard a thousand times, but doesn't it seem odd that in 4+ hours I haven't heard a single song from any of these groups?

¹Anaheim is where the radiation clinic is located. I'm writing this in the waiting room.

The Atlanta Fed was right on target. GDP growth in Q2 clocked in at 2.8%:

There's nothing special to report in the details. Personal consumption was up; investment was up: and government production was up. This is just a very nice, positive report.

On Thursday morning the BEA will announce its first estimate of GDP growth in Q2. Here is the Atlanta Fed's final model-driven forecast:

The blue chip forecast from human beings is 1.8%. By the time most of you read this we should know who got closer.