Month: July 2023

Here are two views of the Palace of Versailles. They show pretty vividly why I've never much liked it: it's one of the most boring piles of stone in Europe. What's more, I've never been a fan of formal French gardens, so I don't even like the gardens of Versailles that much. And I've never been there when the famous fountains were running, so I have no idea how impressive they might be.

As part of a historical narrative, Versailles is fascinating. As a place to visit, though, the truth is that it's mostly a long slog without much of a payoff.

Apparently everyone loves Monica Bertagnolli, Joe Biden's pick to run the NIH. Even Bernie Sanders likes her. But Bernie is miffed at Biden over some trivial issues related to drug pricing, so he's blocking her appointment.

The impasse is just one of several Biden has encountered with respect to his nominees....Biden’s pick to run the Labor Department, Julie Su, has seen her nomination founder for five months amid skepticism from a trio of senators. Nominees for both the Federal Aviation Administration and Federal Communications Commission withdrew this year after it became clear they couldn’t win confirmation. In May alone, Biden lost two judicial nominees. Sen. Tommy Tuberville (R-Ala.), meanwhile, has singlehandedly halted the promotions of around 250 military officers, resulting in vacancies across the Pentagon.

I'm really sick of grandstanding senators who revel in personally grinding routine government business to a halt over their petty beefs and moronic pet peeves. Filibusters are bad enough, but unanimous consent is worse: it makes a running mockery of democracy by allowing a single person to jam things up even if the other 99 want to vote. It's long past due for the Senate to put an end to this idiocy.

Here is a mystery. Ever since the pandemic, Black men have been increasing their participation in the labor force. By March of this year they had increased their participation rate to match white men for the first time in history.

Then, in a single month, participation for Black men suddenly tumbled by 2.7 percentage points and never recovered. Everyone else showed (small) gains that month and ever since.

Unsurprisingly, the unemployment rate shows the same thing. In the past two months unemployment among Black men has suddenly surged from 4.5% to 5.9%. Everyone else has showed no change.

Now, all this does is put Black men back to exactly where they were before the pandemic: an overall labor force participation rate of about 68% and an unemployment rate of about 6%. Still, why the steady gains for three years followed by a plunge in April? What's going on?

In a few days we'll have some wage data for the second quarter, which will tell us if the decline in Black participation has been matched by a decline in Black wages. In the meantime, this is just something to ponder.

More Clarence Thomas weirdness today. His top aide, Rajan Vasisht, was responsible for collecting payments from guests to Thomas's annual Christmas party:

Vasisht’s Venmo account — which was public prior to requesting comment for this article and is no longer — show that he received seven payments in November and December 2019 from lawyers who previously served as Thomas legal clerks. The amount of the payments is not disclosed, but the purpose of each payment is listed as either “Christmas party”, “Thomas Christmas Party”, “CT Christmas Party” or “CT Xmas party”, in an apparent reference to the justice’s initials.

Almost all of the lawyers who made the payments are senior litigators at big law firms.

So . . . former Thomas clerks who are now senior attorneys were apparently being charged admission to Thomas's party. I don't know if that's illegal, but it's a little unusual, isn't it? Do people normally charge guests to come to a holiday party? Corrupt or not, Thomas sure is a weird dude.

As usual, if it's CPI day it's also earnings day. Both weekly and hourly earnings for blue-collar workers were up at an annual rate of 4.7% from June to May. With headline inflation running at 2.2%, this means that worker pay went up about 2.5% in real terms. Not too bad.

Hourly wages for all workers were up only 4.4%, or 2.2% in real terms. However, the number of hours worked ticked up a bit, so the increase in weekly wages was 8.1%, or 5.8% in real terms. That's good news for workers, who are getting a bit more work than before, and probably not a problem for the Fed since the increase is mostly due to longer hours worked, not underlying pay rises.

As usual, it's worth noting that real weekly earnings have gone up a grand total of 0.99% since the beginning of 2020. It's a mystery why the Fed is apparently so worried about spiraling pay worming its way into inflation.

The CPI report for June was released this morning and it's nothing but good news.

Headline inflation is at 2.2% and dropping, while core inflation finally cracked and is now down to 1.9%. Inflation is literally running at the rate the Fed wants it to.

If you insist on wanting to know the conventional year-over-year figures, they were 3.1% for headine CPI and 4.9% for core CPI. But all this does is incorporate high inflation from the second half of last year into this year's lower figures, so why would you do that?

In other news, as foreshadowed last night, if you remove shelter from core inflation then CPI is literally zero:

Inflation news is spectacular this month. We just needed a little patience, that's all.

Over at the Wall Street Journal, James Mackintosh gets some surprising front-page treatment for a dovish take on inflation:

If core inflation came in just below 3%, the Federal Reserve would breathe a huge sigh of relief....It isn’t merely a dream: Measure U.S. price changes the way Europe does, and inflation was already there in May.

....U.S. core inflation—which excludes volatile food and energy—measured using the standard consumer-price index was 2.3 percentage points higher than the European-style inflation, known as the harmonized index of consumer prices. It is the biggest gap there has ever been.

It's actually much more dramatic than that. Mackintosh is using conventional year-over-year figures, which aren't very useful when inflation is changing quickly—as it is now. If, instead, you measure monthly inflation to get a better sense of where things are right now, then core HICP is an astonishing 5.1% lower than core CPI:

Why is US inflation so much lower if you calculate it the European way?

The main reason is that Europe’s measure, known as HICP, doesn’t include the imaginary cost of what a homeowner would pay to rent their house....Exclude something that no one actually pays, and which is calculated from guesses by homeowners of the rental value of their house, and core inflation’s looking basically fine.

The BLS will have new inflation figures for us Wednesday morning, although I don't know how quickly their HICP estimate gets updated. Stay tuned.

Philip Tetlock's Forecasting Research Institute conducted an exercise last year on the topic of human extinction events. Why? Who knows. Their report was released yesterday and, long story short, a group of various experts predicted a roughly 10-20% chance of catastrophe by 2100 (10% of human race dies) and a 1-6% chance of total extinction (fewer than 5,000 humans left). This is the total sum of the risk from five possible scenarios:

• Artificial intelligence
• Nuclear
• Engineered virus
• Natural virus
• Meteor, sun going nova, etc.

Let me say from the start that I think this is kind of nuts. I'd personally put the overall risk of extinction from everything combined at about 0.1% or so. But that's just me. I'm an eternal optimist, I guess.

The proximate motivation for the study, of course, was the recent surge of interest in the possibility of some future AI going nuts and killing everyone. AI experts overall predicted a 13% chance of AIs causing an extinction level event, and I found this breakdown instructive:

Apparently there's a group of AI experts who are absolutely obsessed with extinction and have spent more than 1,000 hours thinking about it. These people seem to have thought themselves into a frenzy and they're obviously driving up the average a lot. The median extinction forecast, which eliminates the influence of the extreme outliers, is only 3%. Still too high, I think, but not completely crazy.

Unlike some things, this is not a case where thinking harder does you any good. There's really no concrete evidence one way or the other about the chances of AI-caused extinction, just speculation, and going down an extinction rabbit hole is little more than a fast path to the booby hatch. Just keep in mind that any possible super-threat—AI or otherwise—will almost certainly be met by a countervailing super-defense. COVID-19 was a dangerous pathogen, but advancing technology allowed us to create a vaccine in record time. Nuclear weapons are being met with missile defense systems. Climate change is being met by renewables and, maybe someday, geoengineering. AIs will be met by other AIs. So chill.

Well, this is weird. A few years ago police investigators found out that four state troopers in Connecticut had been entering fake traffic tickets into an internal police database:

Four troopers had collectively entered at least 636 fake tickets into the state police computer system over a nine-month stretch to make it appear they were more productive than they actually were. The troopers, who worked for Troop E based in Montville, did so for their own personal benefit — to curry favor and perks from supervisors, internal investigators concluded.

Huh. But it gets weirder. A recent audit discovered that upwards of a quarter of the the entire Connecticut state police had been doing the same thing over the past decade. But why? Surely not for personal perks if they were all doing it. The answer, it turns out, is that police in Connecticut are required to report every traffic stop they make to the Connecticut Racial Profiling Prohibition Project, a state-funded group that analyzes police citations to determine racial profiling trends:

The project, which is contracted by the state to analyze the records, including race, ethnicity, gender, age and other demographic information about drivers stopped, to identify patterns of potential racial profiling by officers.

....The audit found with “a high degree of confidence” troopers submitted at least 25,966 tickets to the racial profiling database that did not match records in the judicial database between 2014 and 2021.

....Overreported records were more likely to be reported as white drivers and less likely to be reported as Black or Hispanic drivers, the auditors found.

There had been patterns reported previously that state police were more likely to stop Hispanic motorists during daylight hours and more likely to search drivers of color. Troopers probably found these allegations annoying and wanted investigators to get off their back, so—apparently—they cooked up a scheme to issue fake tickets that had the overall effect of boosting the number of white drivers in the database. This made it look less likely that troopers were engaged in racial profiling. The whole thing went on for years and involved nearly a thousand state police.

Or maybe it was all just a big mistake. Governor Ned Lamont urged caution:

“I wouldn't jump to conclusions,” said Lamont, who became governor in 2019, a few months after state police found four troopers had been fabricating tickets. "There's no indication that was purposeful. A lot of it may have been inadvertent. We've got to look into that.”

Perhaps. I guess anything is possible.

Flyers crossing into the international terminal at LAX.