Apropos of nothing in particular, I got curious this morning about illegal immigration and field workers. About half of all field workers are undocumented, so if there's been a surge of illegal immigration lately, as some have speculated, you'd expect to see the wages of field workers decline. But how would you measure that?

I'm not sure what the best approach is, but I decided to compare the wages of field workers to the wages of all nonsupervisory workers. Here's what I got:

Relative wages for field workers were flat all through the aughts, as illegal immigration was climbing, and declined a bit during the Great Recession. However, since 2012 they've risen three percentage points. In 2016, field workers earned nearly 57 percent of the average nonsupervisory wage.

Based on this, I'm willing to bet that that illegal immigration hasn't surged over the past couple of years. Just the opposite, maybe, which would be consistent with the rise in field worker wages since 2012.

Another week, another pivot gone awry:

For Mr. Trump, this was supposed to be a week of pivoting and message discipline. The president read from a script during public appearances and posted on Twitter less often. He invited lawmakers from both parties to the White House for strategy sessions on the health measure. He scheduled policy speeches, like one near Detroit, where he announced that he was halting fuel economy standards imposed by Mr. Obama.

....But by Friday, as Mr. Trump worked to call attention to his powers of persuasion in securing commitments from a dozen wavering Republicans to back the health measure, the White House was left frantically trying to explain why Mr. Spicer had repeated allegations that the Government Communications Headquarters, the British spy agency, had helped to eavesdrop on the president during the campaign.

There's a piece of me that hardly blames reporters for replaying the "pivot" narrative over and over. Let's face it: It defies human understanding that an easily bored 8-year-old has been elected president of the United States. But he has—and every week he promises to be good. Maybe he even tries. Who knows?

For something like 50 or 60 consecutive weeks, the Trump entourage has been insisting that the boss is going to pivot and start being presidential real soon now. How long before everyone understands it's not going to happen?

We have 3.8 years of this acting out left. It's time for everyone to give up on the fantasy that Trump is going to turn into an adult someday.

Someday I will get bored of regaling you with pictures from my new camera, but that day is not yet. It turns out that burst mode is both a blessing and a curse.

First the curse. With my old camera, I never culled any of the photos I took. I just copied everything onto my desktop PC and left them there. Storage is cheap, and no matter how many pictures I took, I was never going to run out of room.

Then, a few days ago, I ran out of room. Not on the desktop machine, but on my tablet, which is synced to the desktop. For some reason, Microsoft uses MicroSD for expansion memory on the Surface Pro 4, and at the time I got it that meant a maximum size of 128GB. But the new camera takes pictures that are about twice the size of the old Canon, and burst mode means I can crank through a couple of thousand shots in a few days. So the Surface croaked.

In the short term, the answer was a bigger memory chip, which is thankfully available now. In the longer term, it means—what? Going through all the files every time I transfer them and weeding out 90 percent of them? That sounds tedious. Stop syncing some of the folders? I'm tired of that. It's handy to have everything available everywhere at all times. I'll have to ponder this.

But burst mode is a blessing too. I've tried a few times in the past to get a picture of Hilbert scurrying down a tree after climbing onto the roof, but I could never do it. My old camera had terrible shutter lag, slow autofocus, and no burst mode. Unless I timed it perfectly—and I never did—I couldn't get the shot.

But burst mode plus fast autofocus makes it easy! Check it out:

Now we can all pretend to be Vogue editors, choosing just the right shot for this month's cover. I chose No. 3, which shows Hilbert at his graceful best:

So there you have it: better action photography, but lots and lots of gigabytes. Is there a solution that's neither mind-numbingly tedious nor exorbitantly expensive?

Yesterday a reader asked me what was happening with the new plan to annihilate ISIS, which was supposed to be ready to go at the end of February. It's done, I told him, but it hasn't been released yet. He wrote back, asking how I knew stuff like this. I told him my secret source: I think I read about it in the New York Times.

All well and good, but what's in the plan? My secret source this time is NBC News:

Now, the Pentagon has given [Trump] a secret plan, but it turns out to be a little more than an "intensification" of the same slow and steady approach that Trump derided under the Obama administration, two senior officials who have reviewed the document told NBC News.

The plan calls for continued bombing; beefing up support and assistance to local forces to retake its Iraqi stronghold Mosul and ultimately the ISIS capital of Raqqa in Syria; drying up ISIS's sources of income; and stabilizing the areas retaken from ISIS, the officials say.

Gee, I thought we were supposed to be bombing the shit out of ISIS and taking Iraq's oil, but apparently that plan got lost somewhere between Election Day and now. Or did it? After all, there's no chance that President Trump is going to announce this new plan as an "intensification." He's going to go on TV and claim that it's the biggest military operation since D-Day. It makes Rolling Thunder look like kids with popguns. It's more strategically brilliant than the Inchon landing. And it will be a more famous victory than even the Gipper's invasion of Granada.

When you hear all this stuff, just remember that it's Trump's usual "truthful hyperbole." In reality, the new operation is just going to be a modest uptick in the tempo of the Obama plan that's been gradually and steadily making progress for the past two years.

Lunchtime Photo

Here's another long-exposure shot of a freeway. It's better! I found an overpass (Bear St.) with no fencing, so that eliminated the blurry diagonal stripes. I got there at the right time, which produced a more interesting picture. And the overpass was less shaky and allowed the use of a tripod since there was no fencing in the way.

Who knows? Maybe #3 will be even better when I get around to it.

One of the surprising things about the CBO score of AHCA, the Republican health care bill, is their conclusion that premiums will fall starting in 2020. By 2026, average premiums will be 10 percent lower than they would be under Obamacare. But why? Here's CBO:

First, the mix of people enrolled in coverage obtained in the nongroup market is anticipated to be younger, on average, than the mix under current law. Second, premiums, on average, are estimated to fall because of the elimination of actuarial value requirements, which would result in plans that cover a lower share of health care costs, on average.

....By 2026, CBO and JCT project, premiums in the nongroup market would be 20 percent to 25 percent lower for a 21-year-old and 8 percent to 10 percent lower for a 40-year-old—but 20 percent to 25 percent higher for a 64-year-old.

Hmmm. Let's translate this into English. First, CBO assumes that premiums will go up for old people, forcing many of them to drop out of the market. Since old people have expensive premiums, fewer old people means the average for the remaining pool will be lower.1 Second, AHCA policies will cover far less of your medical expenses, so naturally they'll be cheaper.

The chart below shows how this "reduces" average premiums. If you use CBO's projections and do a little arithmetic assuming a modestly younger pool, you get the average premium estimate for the overall pool shown on the left. AHCA is cheaper than Obamacare.

But the current age breakdown in the Obamacare insurance pool is 28 percent young, 38 percent middle-aged, and 26 percent old. What if you assume that stays the same? You get the premium estimates in the middle.

Finally, what if you assume that AHCA paid for 87 percent of your medical bills, just like Obamacare? Then you get the premium estimate on the right.

In other words, if you compare apples to apples, AHCA produces far higher overall premiums than Obamacare.2

Note that CBO didn't do anything wrong here. They simply did their projections based on a (correct) assumption that AHCA would be too expensive for many old people and would produce crappier policies that had higher deductibles and paid far less of your medical bills. The "average" premium is lower, but obviously not in a way that helps anybody in real life.3

1Think about it this way. If a high school sends all its A students to a magnet school across town, the school's average GPA will go down. This is despite the fact that nobody's grades have actually changed.

2This is a fairly extreme example because the actuarial value changes a lot (87 percent vs. 65 percent) for the cheaper policies preferred by low-income folks. CBO has a second example that uses a middle-class worker, and it produces similar but less dramatic results.

3Hardly anybody, that is. If you're young and don't get any medical care, then the lower premiums really do help you.

Before everyone goes too far down the rabbit hole on President Trump's "Obama had my wires tapped" tweet, I just want to remind everyone that this is what he said:

We've known almost from the start—because White House sources have confirmed it—that Trump based his tweet on a summary of a Mark Levin radio rant published in Breitbart News on March 3. The article is a timeline of "the known steps taken by President Barack Obama’s administration in its last months to undermine Donald Trump’s presidential campaign." Here's the bit that Trump was almost certainly talking about:

4. October: FISA request. The Obama administration submits a new, narrow request to the FISA court, now focused on a computer server in Trump Tower suspected of links to Russian banks. No evidence is found — but the wiretaps continue, ostensibly for national security reasons, Andrew McCarthy at National Review later notes. The Obama administration is now monitoring an opposing presidential campaign using the high-tech surveillance powers of the federal intelligence services.

Trump had "just found out" about this because he doesn't read Slate, where the story of the suspicious server was first published, and the rest of the media didn't give it much coverage. The "Obama" part is Trump's typical attack dog mode. The FBI investigated the server but found no evidence of wrongdoing, and Obama himself was never involved. "Trump Tower" is because that's where the suspicious server was. And the Slate piece ran on October 31, "just before the victory."

Since Trump's tweet first burst on the world, the White House communications team has feverishly looked for news sources other than Breitbart that they can credibly name as the source of Trump's tweet. But none of them fit. The DNC hacks had nothing to do with Trump Tower. The Trump "dossier" had nothing to do with surveillance. The general investigation of Trump's aides and their ties to Russia has been ongoing since summer, and Trump had known about it for months. He didn't "just" find out about it. Finally, British news outlets ran stories claiming that the FBI had asked for a FISA warrant to examine Trump aides, but this wouldn't involve Obama personally even if it did happen—and Trump made it clear in multiple tweets that he was accusing Obama personally of ordering the surveillance.

Bottom line: Trump is talking about the FBI's brief investigation of a server in Trump Tower that communicated with a Russian bank. He read about it in Breitbart News the day before his tweet. That's it. He has no evidence that Obama had anything to do with it; that it involved any surveillance of him personally; or, for that matter, that it involved any actual surveillance at all. Just a Breitbart piece based on a Mark Levin harangue based on a British story based on "two separate sources with links to the counter-intelligence community" that has never been confirmed by any of the legions of extremely sharp national security reporters based in the United States.

There is nothing more to this.

I've been so busy taking pictures of other stuff that I haven't had time this week to take very many cat pictures. But here's Hilbert sitting by the door pondering whether to go out into the backyard. He always does, but he often has to think about it long and hard before making his move.



Great Britain:


I'll bet the Rt Hon Chris Patten is required by his employment contract to use the Oxford comma.

Yesterday a reader sent me an email asking, "Is there a more 'you' story in the news right now than this"? The story in question was a court decision that hinged on a Maine law that lacked an Oxford comma, but I got busy with other stuff and never read it. So here it is:

What ensued in the United States Court of Appeals for the First Circuit, and in a 29-page court decision handed down on Monday, was an exercise in high-stakes grammar pedantry that could cost a dairy company in Portland, Me., an estimated $10 million.

....The debate over commas is often a pretty inconsequential one, but it was anything but for the truck drivers. Note the lack of Oxford comma — also known as the serial comma — in the following state law, which says overtime rules do not apply to:

The canning, processing, preserving, freezing, drying, marketing, storing, packing for shipment or distribution of:

(1) Agricultural produce;
(2) Meat and fish products; and
(3) Perishable foods.

Does the law intend to exempt the distribution of the three categories that follow, or does it mean to exempt packing for the shipping or distribution of them?

Shockingly, this lack of an Oxford comma wasn't just sloppiness. Apparently the Maine Legislative Drafting Manual is very clear that Oxford commas shouldn't be used except in rare cases. What idiots! If there's anyplace in the world that ought to embrace the Oxford comma, it's the legislative process. Legislation is mostly impenetrable anyway, so even if you think the Oxford comma is ugly, who cares?

In any case, I'd like to point out that a surprising number of court cases hinge on "grammar pedantry." Or syntax pedantry or dictionary pedantry or various other kinds of linguistic pedantry. Is a "penalty" enforced through the tax code a fine, which would be illegal, or a tax, which would be legal? One Supreme Court justice changed his mind on this crucial question in 2012, and saved Obamacare. Millions of people now have health insurance because of linguistic pedantry.

And before you ask, I am pleased to report that Mother Jones officially endorses the Oxford comma in its style guide. Everyone else should too.