Skip to content

Ben Casselman writes about the economy in the New York Times today:

After five years of uncertainty and turmoil, the U.S. economy is ending 2024 in arguably its most stable condition since the start of the coronavirus pandemic. Inflation has cooled. Unemployment is low. The Federal Reserve is cutting interest rates. The recession that many forecasters once warned was inevitable hasn’t materialized.

I'm not picking on anyone in particular here, but this narrative is just way too popular. We have no idea if the economy is stable right now. It's cooled down since its peak, but that means nothing. It could either stabilize or continue cooling, and no one knows which it will do.

In order to see what I mean, here's a look at real GDP growth, the most common measure of overall economic healthiness. The gray bars show recessions. I've added the red bars to highlight growth a few months before each one:

Almost all of the pre-recession growth trends look pretty stable. They predicted nothing. Now here's a different look:

The green bars show various declines in growth that lasted a year or so. None of them predicted an imminent recession.

We just don't know. Current growth is down a little bit but not a lot. It literally tells us nothing about where we are in the economic cycle.

On the other hand, there are some historically reliable recession predictors that are flashing red—or at least pink:

  1. Job growth has slowed almost to zero over the past year.
  2. Unemployment is up and has triggered the Sahm Rule.
  3. The yield curve just recently reverted from negative to positive.
  4. Nominal interest rates remain high.
  5. New manufacturing orders have been steadily declining.
  6. New housing starts and sales are both sluggish.
  7. The stock market is way overvalued—though admittedly this has no predictive history. It just worries me.

Not everything is bad. Consumer spending remains steady; oil prices are contained; and fixed investment is still growing. Bottom line: I don't know anything more than anyone else, but it's worth at least acknowledging both the good and the bad.

From CNN:

The House Ethics Committee secretly voted earlier this month to release its report into the conduct of former Rep. Matt Gaetz before the end of this Congress, according to multiple sources with knowledge of the matter.

....The vote, which has not previously been reported, amounts to a stark reversal for the panel after it had voted along party lines in late November not to release the results of the investigation. The decision to release the report suggests that some Republicans ultimately decided to side with Democrats on the matter.

Now that Gaetz no longer matters to Donald Trump, I guess at least a few Republicans screwed up the courage to do the right thing.

Even though I have some views on this, I'm weary of the endless debate about what liberals should do to win back American voters. I read the same polls as everyone else, and that's really all of us have to go on. The truth is that unless you have a remarkably large and diverse set of personal friends (you don't), your personal inclinations are hopelessly parochial. My only advice is not to kid yourself with selective poll readings. Suck it up and take the good with the bad.

However, I do have plenty of personal opinions, some based on my own moral compass and others informed by the evidence—and I'm happy to share them. Some would probably be good for the liberal movement and others wouldn't, but at least they aren't just guesses. Here is Shorter Kevin Drum:

Universal health care (strongly for it); illegal immigration (in favor of E-Verify and moderate toughening in general); trans issues (wary of gender-affirming care for teens but opposed to bans); globalization (still for it); China hawkery (on the fence); abortion rights (radically supportive); wokeness (hearts are in the right place but ditch the performative silliness); education (Black-white gap is disgraceful); Gaza etc. (on Israel's side with ever increasing reservations); religion (non-aggressive atheist); death penalty (slightly against); housing crisis (basically just a California problem); YIMBY (on the fence); guns (opposed but accept there's little we can do); climate change (of course it's real and serious); nuclear (mostly supportive); crime (not a growing problem); voter ID (sure, but first make national ID free and mandatory); prison sentencing (way out of control); safety net (a huge liberal success that's both popular and in little danger); long-term nursing care (pay 100% via Medicare); deficit/taxes (no choice but to raise taxes moderately on everyone); terrorism (will mostly die on its own over the next decade); AI (we will have legit AGI by 2033 and, yes, it will take over most of our jobs by 2040-50); democracy (not truly in any lasting trouble); cost of college (mostly a fake issue); affirmative action (prefer class-based preferences); Fox News (a cancer on the country, needs to be destroyed and the earth salted behind it); free speech (pretty close to absolutist, always ask "who's going to do it" if someone proposes an exception); COVID vaccines (100% safe and effective); permitting reform (probably needed, but don't throw the baby out with the bathwater); Joe Biden (still think he was a pretty good president overall); government effectiveness (DOGE is a stupid sham, Congress needs to fix the rules).

Have I missed any big ones?

The number of housing units under construction fell again in November, but it's worth noting that total construction is still higher than it was at the peak of the housing bubble: 1.43 million units vs. 1.42 million units on an annual basis. However, the composition of construction has changed dramatically:

We never recovered our outsize taste for single-family homes, but the number of apartment units under construction has doubled.¹ Two decades ago we built half as many apartment units as single-family homes; today we build more than twice as many:

¹Note that these are units in large-ish buildings of five or more units. Our taste for small apartment buildings has been modest for a long time and hasn't changed much.

Why did the murder rate spike in 2020? There are two obvious possibilities: the killing of George Floyd and the start of the pandemic. The problem is that both of these things happened at practically the same time, which makes it all but impossible to tease out which one was the actual cause.

I took a crack at it anyway a year ago and tentatively decided it was probably the George Floyd killing that did it. But it was iffy. You'd need murder rates almost down to the week to distinguish between the pandemic (mid-March) and Floyd (late May).

Well, a pair of researchers at Brookings extracted exactly that data and came up with this chart:

The red line is the key one. It shows homicides starting to rise in April, before the George Floyd killing, and then continuing to rise at the same rate afterward. The murder rate then stayed high for the rest of the year.

Now, there's no way to get around the fact that there just aren't many data points here. This could easily be a coincidence. So the authors took a deep dive into the national data and came up with this:

Data show that just before the national spike in homicides began, U.S. cities experienced two shocks:

  • Large numbers of teen boys in neighborhoods with concentrated poverty were pushed out of school in March of 2020.
  • Large numbers of young men in neighborhoods with concentrated poverty also lost their jobs in March and early April of 2020.

These two developments had a massive impact on neighborhoods with concentrated poverty. By mid-April, young men laid off due to the pandemic and high school boys whose schools were shut down made up around 1 out of every 30 people living in low-income neighborhoods.

Long story short, young men lost their jobs and were sent home from school. Then in mid-April, according to cell phone data, people started leaving home more often as the COVID lockdowns eased. So you ended up with a whole lot of poor young men roaming the streets with little to do. And according to the authors, cities with larger job and school shocks also had larger spikes in murder:

I'm not 100% convinced by this, but it's an impressive bit of data mining and I'd call it extremely plausible. In truth, it never made a lot of sense that the George Floyd killing could be responsible for a two-year murder spike. It was probably COVID all along, perhaps with just a bit of a tail wind from George Floyd.

RFK Jr. is pandering and kowtowing until his lips turn blue in a desperate attempt to win Republican approval for his nomination to run HHS:

Too many abortions! We can't be the moral leader of the free world with such high abortion rates! Here's the truth about that:

Among all countries, the US ranks 106th out of 151. Pretty low! As for moral leadership of the free world, that would apparently be Algeria, with an abortion rate of 0.4. Definitely a country to look up to.

Elon Musk's latest whinge is that the Biden administration is out to get him. In particular:

  1. This year Reuters won a Pulitzer Prize for its investigation into the seamy side of Musk's businesses.
  2. AT THE SAME TIME, Biden has awarded $300 million in federal contracts to Reuters!!!

I wouldn't normally waste time with this, but I am very, very bored sitting in my hospital bed waiting for my final dose of Tecvayli. So I got curious. As best I can tell, the story originated here and got picked up by Musk here:

WEAPONIZATION: Biden publicly ordered 11 federal agencies to 'look' into Elon Musk. All 11 opened investigations. Then all 11 funneled a total of $300M to Reuters who then who won a Pulitzer Prize for their relentless attack pieces against the billionaire.

Wow. Or, as Musk himself eloquently puts it:🤨

Anyway, not to keep you in suspense, but yes, the federal government has a lot of contracts with Reuters. More specifically, they are almost all with Thomson Reuters, a provider of commercial information, of which the Reuters news network is a small part. Luckily, USAspending.gov tells us the size and award date of every Thomson Reuters contract and handily presents it by month:

There you have it: $297 million in Reuters contracts under Joe Biden!!!

Compared to $449 million under Donald Trump, which, among many other things, includes about $60 million to ICE for help tracking down illegal immigrants via their CLEAR database.

This is Elon Musk these days. He's either a complete loon or else he just doesn't care anymore. Even Donald Trump is a little more careful about what he retweets.

Just to get serious for a moment, there's really no mystery about New Jersey's mystery drones. Most of them are just ordinary airplanes and whatnot, while the rest are . . .

drones.

Right? I mean, what else could they be? Even if, for the sake of discussion, there are lots of them around, they're still drones. So why the freakout?