Month: October 2023

For those of you who missed it over the weekend, here's a chart showing the total population of unauthorized immigrants in the US over the past 15 years:

This chart is based on consensus estimates of the unauthorized immigrant population. The final two years are my own (possibly iffy) projections based on the number of border encounters in 2022 and 2023.

Long story short: for all the hysteria over mobs of unauthorized immigrants on the southern border, their actual population in the US has barely changed at all. It's been hovering around 11 million for 15 years and so far shows little sign of changing.

It's the first Monday in October, which is a good excuse to post this late afternoon picture of the Supreme Court building. I'm sure that everyone inside is eager to get back to work fighting for justice.

Donald Trump's trial for business fraud has opened on schedule. The New York Times provides a summary of the action so far:

In their respective opening statements, the attorney general’s office and lawyers for Donald J. Trump spoke past each other: While the attorney general’s lawyer focused on the specific mechanics by which properties were valued — and why — Mr. Trump’s lawyers continued to argue that, overall, there had been nothing wrong with the former president’s financial statements.

Kevin Wallace, a lawyer for the attorney general’s office, told the courtroom that employees of Mr. Trump had reverse-engineered the value of individual assets — properties like Trump Tower and 40 Wall Street — to arrive at the former president’s desired net worth. He played a clip of Michael D. Cohen, Mr. Trump’s former fixer, explaining the process, prompting the former president to cross his arms and shake his head, scowling.

In other words, Trump's lawyers are barely even bothering to make a case. Remember that this is solely the penalty phase of the proceedings. The trial judge has already decided that Trump illegally inflated the value of his properties, so that's not at issue. There's no point in continuing to argue about it except as fodder for an appeal, which appears to be all they're doing right now.

In related news, Trump headed toward the TV cameras at the break and called for judge Arthur Engoron to be disbarred. “This is a judge that should be out of office,” he said. “This is a judge that some people say could be charged criminally for what he’s doing.”

"Some people." Indeed.

In the New York Times today, Margaret Renkl rails against book banning:

During the 2022-2023 school year, PEN America, an advocacy organization that defends free expression, recorded 3,362 instances of book banning, a number that represents an increase of 33 percent in just one year....Book bans belong to the same categorical crime against democracy as denying red-state citizens the full range of medical care available to the citizens of blue states.

Maybe I shouldn't care about this use of language, but increasingly I do. We are not talking about book bans here. All of the books in question continue to be freely available at public libraries, bookstores, and online. All that's happening is that school libraries are deciding what books are appropriate for schoolchildren, and many of them are making judgments that we liberals don't like.

Hell, I don't like them either. Nevertheless, this isn't a "crime against democracy." It's just a misguided panic instigated by a small number of people in a small number of mostly rural, conservative school districts.

So what should we call this if it's not really a ban? A squeeze, because it reduces the number of places kids can find some of these books? A restriction? An exclusion? A suppression? I don't know. But until we're genuinely banning books, as we used to in the days of Ulysses and Tropic of Cancer, we really ought to find something else to call it.

POSTSCRIPT: For what it's worth, PEN America says that about 1,400 books were added to its index in 2023 through June, of which 60% are “banned pending investigation.” In other words, they were temporarily removed while a challenge is considered. The number of books that have been removed permanently is considerably smaller.

The Wall Street Journal says American consumers are spending lots of money on big blowouts:

Household spending, the primary driver of the nation’s economic growth, remains robust. Americans spent 5.8% more in August than a year earlier, well outstripping less than 4% inflation.

....Economists and financial advisers say consumers putting short-term needs and goals above long-term ones is normal. Still, this moment is different, they say.

A tough housing market has more consumers writing off something they’d historically save for, while the pandemic showed the instability of any long-term plans related to health, work or day-to-day life. So, they are spending on once-in-a-lifetime experiences because they worry they may not be able to do them later.

I'd like to fact check this, but how? The Journal provides zero evidence that "once-in-a-lifetime" spending has increased, and overall consumer spending is perfectly normal right now:

But maybe these aren't normal times?

“Normally at a time when you have higher inflation, but also higher interest rates, you don’t expect spending to hold up so well,” says Wilbert van der Klaauw, an economic research adviser on household and public policy at the Fed.

The thing is, that's not true:

Consumer spending, even adjusted for inflation, never declines outside of recessions—and only occasionally during recessions. Over the past 30 years it's happened exactly twice: first during during the Great Recession and then, briefly, at the start of the pandemic.

So I dunno. Are Americans increasingly splurging on big blowouts they can't afford? Maybe. But is there any actual evidence for this?

Celebrity author Michael Lewis spent months shadowing Sam Bankman-Fried, founder of the failed crypto exchange FTX, and apparently came away convinced that SBF was just the victim of misfortune:

This isn't a Ponzi scheme. Like, when you think of a Ponzi scheme, I don't know, Bernie Madoff, the problem is— there's no real business there. The dollar coming in is being used to pay the dollar going out. And in this case, they actually had-- a great real business. If no one had ever cast aspersions on the business, if there hadn't been a run on customer deposits, they'd still be sitting there making tons of money.

WTF? It's true that FTX wasn't a Ponzi scheme. After FTX's bankruptcy, John Ray was appointed to clean up the mess and he called it correctly:

New FTX head John Ray III is asked to compare FTX to Enron. “[Enron crimes] were highly orchestrated financial machinations by highly sophisticated people to keep transactions off balance sheets. This (FTX)... isn't sophisticated whatsoever, this is just plain old embezzlement."

— Paul McLeod (@pdmcleod) December 13, 2022

"Plain old embezzlement." In a nutshell, here's what happened: one of SBF's companies (Alameda Research) borrowed $10 billion from another of SBF's companies (FTX) and then "loaned" SBF $3 billion. In other words, garden variety embezzlement. And that's not even counting the fact that the initial loans were wrong in the first place. The Alameda loan depleted FTX's treasury, which was supposed to be just a boring guardian of funds that its customers were using for trades. When customers found out it had lost $10 billion, they started asking for their money back and FTX didn't have it. This wasn't some irrational, panicky run. It was completely sensible based on the fraud that was obviously going on.

In the end, what doomed FTX was the same thing that doomed so many other crypto enterprises: They accumulate vast sums of money by promising stratospheric returns to investors, but there's no way they can make good on that. As reality starts to close in, they start making riskier and riskier market bets in the desperate hope that maybe one of them will pay off.¹ But they never do, and the business finally goes up in smoke.

But not before the founders make off with a billion or three. This is the story of dozens of big crypto firms. The whole industry is a scam based on the "greater fool" theory: crypto itself may have no value, but you'll be OK as long as you buy low and then find a fool willing to pay more for your tokens before the whole thing collapses. It's made up out of thin air, and to thin air it eventually returns.

¹This is the story of Alameda Research, which was a hedge fund. The twist is that as their risky investments failed they borrowed $10 billion from FTX and then lost that. So both companies ended up bankrupt.

I mentioned the other day that I was curious about the net illegal immigration rate into the US. This would be (roughly) the number of border crossings minus repatriations and people going home of their own accord.

The Border Patrol doesn't provide anything like this, but there's a way of getting at it. Several organizations estimate the total size of the unauthorized immigrant population in the US, and the Migration Policy Institute recently gathered together the data through 2021 here. I took the average of the estimates and then plotted it against border encounters:

The total population of unauthorized immigrants barely changes at all in response to border crossings. In 2021, for example, the Border Patrol recorded 2 million border crossings but MPI estimates that the population of unauthorized immigrants increased by only 200,000. If you extrapolate this to 2023, we will end the year with a population of 11.3 million unauthorized immigrants, barely more than the average of the past 15 years.

Now, this extrapolation is a little dicey, but it really does seem as if net unauthorized immigration is close to zero thanks to repatriations, deportations, and voluntary departures. This, in turn, suggests that the level of illegal immigration isn't really tied much to national policy at all. Rather, it responds thermostatically to the demand for immigrant workers, going up when immigrant departures are high and going down when immigrant departures are low.

Long story short, American businesses want about 11 million unauthorized workers, and as long as that demand is there the population of unauthorized workers will hover around 11 million. The incoming flow of unauthorized immigration merely responds to changes in that population.

POSTSCRIPT: My theory will get a torture test when we get population data for 2022. Will it turn out that the population of unauthorized immigrants stays fairly steady in the face of a second year of 2+ million border crossings? Come back in a year and find out.

Yesterday Rep. Jamaal Bowman triggered a fire alarm in the Cannon Office Building. He says it was because he was heading out to the Capitol Building to vote on the continuing resolution to keep the government open. The door he used is normally open, but on Saturday it was locked and had a confusing sign posted. He thought he needed to pull the alarm to open the door.

Republicans have a different story. They say Bowman did it because he wanted to trigger an alarm that would delay the vote. Republican House members say this was outrageous behavior and some have called for his resignation.

I believe Bowman. The reason is this picture:

The picture was released by Republicans on the House Administration Committee, but the full surveillance video wasn't. This is classic Republican ratfuckery: release some small snippet of evidence that looks suspicious and gets splashed all over the news, but is later found to be benign when the entire context becomes public.

There's no reason to believe anything Republicans say about this unless they release the entire video. If and when they do, I'll lay long odds that it shows Bowman acting a bit confused, then pulling the alarm and leaving the building.

Assuming, of course, that Republicans ever let us see the surveillance video. Which they probably won't.

Great moments in financial analysis from Tom Bevan:

Academics & economists: We can't understand why people don't think the economy is great.

WSJ: Rise in prices from Dec 2019 to Aug 2023:

Gallon of milk: +12.5%
New car: +29.9%
Dozen eggs: +32.5%
Gasoline: +49.4%

Let's expand our view a bit:

If you cherry pick and don't account for wage growth, you can make anything look terrible. In reality, overall prices have increased 18.4% since the end of 2019¹ while overall wages have increased 19-23%. That's no great shakes for American workers, but for most people the relative cost of things has barely changed at all over the past four years.

¹At most. Prices have increased only 16.0% according to the PCE price index.

Yesterday was my last day on Obamacare. As of October 1, I'm officially a Medicare beneficiary.

In other words, I am old.