Kevin Drum

Why is it so hot? The latest theory making the rounds is that it's due to a change in shipping regulations. In 2020 the International Maritime Organization reduced the allowable sulphur content of bunker fuel to 0.5%, which was great for human health but not so great for global warming. You see, sulphur emissions cool the climate, so if you reduce them the climate warms up.

But don't be too quick to believe this. Here's a bit of history followed by some estimates of what the new rules mean for climate change.

In 1997 the IMO adopted what's known as Annex VI, which took effect in 2005. It set a maximum sulphur content of 4.5%. In 2012 this went down to 3.5% and in 2020 to 0.5%. However, SO₂ emissions didn't respond strongly in every case:

A steady increase in shipping overwhelmed the 2005 limit, and the 3.5% limit had only a modest effect. However, the 0.5% limit in 2020 made a big difference.

But what effect did this have on global warming? For that we need to compare two charts from a 2009 analysis of sulphur limits. Pay attention to the green lines:

The chart on the left shows the climate effect if emissions had stayed at year 2000 levels. The chart on the right shows the effect if emissions are cut by 90%.

After 50 years, global temps go up about .051°C if nothing is done. Temps go up only 0.0058°C if sulphur emissions are cut. This comes to a difference of 0.045°C. That's four-hundredths of a degree.

But it hasn't been 50 years since the new limits were set. It's been three years. The net effect on global temps so far has been less than 0.02°C, which is lower than the normal increase for a single year. In comparison, global temps in 2023 are 0.16°C higher than last year, so sulphur emissions are—at most—responsible for only one-eighth of the increase. The other seven-eighths are likely due to the eruption of a huge undersea volcano near Tonga last year and an especially strong El Niño this year.

Bottom line: the new sulphur rules do have an effect, but so far a pretty small one. They've played only a tiny role in this year's long hot summer.

The New York Times tells us today about yet more billionaire assistance propping up the not-so-down-to-earth lifestyle of Supreme Court justice Clarence Thomas. In 1999 he bought a used RV that he uses to travel around the country:

His Prevost Marathon cost $267,230, according to title history records obtained by The New York Times. And Justice Thomas, who in the ensuing years would tell friends how he had scrimped and saved to afford the motor coach, did not buy it on his own. In fact, the purchase was underwritten, at least in part, by Anthony Welters, a close friend who made his fortune in the health care industry.

....Roughly nine years later, “the loan was satisfied,” Mr. Welters added....He would not say how much he had lent Justice Thomas, how much the justice had repaid and whether any of the debt had been forgiven or otherwise discharged. He declined to provide The Times with a copy of a loan agreement — or even say if one existed. Nor would he share the basic terms of the loan, such as what, if any, interest rate had been charged or whether Justice Thomas had adhered to an agreed-upon repayment schedule.

It's almost pointless to say this, but apparently Supreme Court justices don't have to disclose vehicle loans, so there's nothing technically illegal about this. Assuming, of course, that the loan was made at market rates and was fully paid back, which neither Welters nor Thomas is willing to prove.

I wish I were a Supreme Court justice. Sounds like a nice gig.

The chart below shows the net number of people added to corporate payrolls each month in the US. As you can see, it's been going steadily down for the past two years:

I let Excel draw two trendlines through the raw data. If things level off, we get a soft landing. If a straight line is a better predictor, we get to zero by the start of 2024 and then go negative.

There's not really any strong reason to prefer one trendline over the other. They're both equally good fits. It all depends on whether you think the Fed's rate hikes (a) have already done their work and are fading away, or (b) will start to have their biggest impact later this year.

I think it's the latter, but who knows? Which one we get is sure going to make a big difference to next year's election, though.

This is Charlie in his favorite lizard hunting grounds. It's the west side of our backyard, which is full of ferns and other low-growing little plants, which makes it a great place to slink around and pretend no one can see you.

A recent study in Sweden randomly assigned breast cancer screenings to either an AI or a human radiologist. Guess which one did better?

There were 244 screen-detected cancers...in the intervention group and 203 cancers...in the control group. Compared with the control group, the screen-reading workload of radiologists was reduced by 44.3% using AI-assisted techniques.

That's a 20.2% improvement in cancer detection with only half as much work from radiologists. In blunter terms, it means that if AI were used for everyone a whole bunch of radiologists could be let go—and those remaining would get paid a lot less.

I don't imagine this will be allowed to happen without a fight. If AI improves and further testing results are positive, hospitals will be eager to use it. The American Board of Radiology might be less thrilled.

The Fifth Circuit Court is the most conservative in the country. But sometimes you get lucky when a three-judge panel is randomly chosen to hear your appeal. That happened today in a ruling handed down about Section 241 of Mississippi's state constitution, which disenfranchises anyone convicted of a felony—forever. This case ended up being heard by three old judges appointed by Carter, Reagan, and Clinton, not young firebrand conservatives appointed by the likes of George W. Bush or Donald Trump.

As a result, a majority held, first, that Section 241, which was written in 1890, had explicitly racist underpinnnings. They recounted this history in some detail before pointing out that it continues to be applied disproportionately to Black people:

Sections 241 and 253 continue to be part of the Mississippi Constitution and over the years they have been remarkably effective in achieving their original, racially discriminatory aim. In 2017, 36% of voting-age citizens in Mississippi were black. Yet...of the nearly 29,000 Mississippians who were convicted of disenfranchising offenses and have completed all terms of their sentences between 1994 and 2017, 58%—or more than 17,000 individuals—were black. Only 36% were white.

Second, they ruled that Section 241 had no legitimate rehabilitative, deterrent, or retributive function. Also, crucially, a large majority of states have decided that contemporary mores don't allow permanent and widespread disenfranchisement:

“No right is more precious in a free country” than the right to vote....Mississippi denies this precious right to a large class of its citizens, automatically, mechanically, and with no thought given to whether it is proportionate as punishment for an amorphous and partial list of crimes. In so excluding former offenders from a basic aspect of democratic life, often long after their sentences have been served, Mississippi inflicts a disproportionate punishment that has been rejected by a majority of the states and, in the independent judgment of this court informed by our precedents, is at odds with society’s evolving standards of decency. Section 241 therefore exacts a cruel and unusual punishment on Plaintiffs.

There is, of course, a downside to getting lucky judges: If the case is reheard by the entire court, it might well get reversed. Stay tuned.

If you ask ten people to define fascism, you will typically get zero real answers. This is because fascism only ever existed once and that was 80 years ago. What's more, most people have no idea what even the original Italian fascists were about, let alone the hazy variety that's been a favorite insult since the '60s.

Like "neoliberal," which these days just means bad and left of center, fascist is now just a generic term for bad and conservative. It has lost all meaning and nobody should use it anymore. If you can't figure out how to say what you really mean without it, maybe you should rethink what you really mean.

The American economy gained a meager 187,000 jobs last month. We need 90,000 new jobs just to keep up with population growth, which means that net job growth clocked in at an even more meager 97,000 jobs. The headline unemployment rate ticked down slightly to 3.5%.

Overall, this is a weak but not catastrophic jobs report. There are no gotchas in the details, and the employment level increased by about 250,000, very close to what the jobs report says.

In somewhat better news, weekly blue-collar earnings increased 3.0% after adjusting for inflation. This is the third month in a row that they've gone up:

This is bad news for the Fed, which is obsessed with labor costs, but good news for workers.

For reasons best left unexplored, I got curious about long-term suicide rates last night and ended up going down a rabbit hole. What I discovered is that everyone is doing it wrong.

Most of the charts you see are based on reports written decades ago plus CDC figures for the past couple of decades. The problem is that these use different methodologies and aren't comparable. However, in 2009 some researchers at the National Center for Health Statistics went back to the original worksheets for death rates and recalculated everything using the current standard. This is available in a series of reports for 1900-1998, while the CDC's WONDER database (as usual) can be used to fill in the figures for 1999-2021. Here it is:

After two decades of slowly rising, the US suicide rate has stabilized over the past few years. It is now at the same level as the 1950s.

UPDATE: The original version of this post had a mistake in recent suicide rates. They haven't gone down, only stabilized. The chart and the text have been corrected.

POSTSCRIPT: Here is the specific methodology used:

1900-1998: Age-adjusted death rates were calculated using the year 2000 standard using the following source documents: age-specific death rates from unpublished worktables from CDC/National Center for Health Statistics, National Vital Statistics System, Mortality data; data from the Special Reports (1956) publication; and electronic data tapes.

1999-2021: Data were analyzed using National Vital Statistics System multiple cause-of-death mortality files. Suicide deaths were identified using International Classification of Diseases, 10th Revision underlying cause-of-death codes U03, X60–X84, and Y87.0. Age-adjusted death rates were calculated using the direct method and the 2000 U.S. standard population.

A lot of people, either accidentally or deliberately, have misstated a few things about Wednesday's indictment of Donald Trump for colluding with others to overturn the 2020 election results and keep himself in power. Here they are:

1. Free speech. Even mob bosses have free speech rights. But if your local mafia don tells one of his stooges to kill someone—and you can prove it—then he can be put on trial for murder even if he's not the one who pulled the trigger. Likewise, the Trump indictment is not about speech. It's about planning and conspiracy.
2. Insurrection. There is nothing in the indictment about Trump being responsible for the mob violence on January 6. Nor is there anything about insurrection. He is not being tried for that.
3. Lying. The prosecution doesn't have to prove that Trump knew he was lying when he conspired to overturn the election results. The standard of proof stops short of requiring mind reading: "Reckless disregard of whether a statement is true, or a conscious effort to avoid learning the truth, can be construed as acting 'knowingly.' "

Everybody got that?