Six years ago I read a pair of articles about Yemen which predicted that its population would double by 2035; oil revenue would decline to zero by 2017; and the capital city of Sanaa would run out of water by 2015. Today I got curious: How are those forecasts panning out?

Population: On target. Yemen’s population has increased from 23.6 million to 27.5 million since 2010—an annual growth rate of 2.58 percent. If this continues, Yemen’s population will double by 2037.

Oil revenue: On target. Yemen is currently producing a meager 22,000 barrels of oil daily. In fairness, much of this is due not to pumping their fields literally dry, but to infrastructure destruction during the current civil war. They still have proven reserves of about 3 billion barrels, so production could rise again if the war ever ends.

Water: On target? Adela Jones of USC writes: “Already, Yemenis allocate up to 30% of their annual income towards water….As early as 2017, Sana’a may officially run out of water. Given consumption trends, the rest of the nation may follow.”

I remain fairly ignorant about Yemen, aside from the fact that it’s the site of a brutal proxy war between Saudi Arabia and Iran—in Saudi Arabia’s view, anyway—and we’ve been assisting the Saudis since it started. But Yemen’s future looks pretty bleak no matter who wins. What happens when they finally pump the last of the groundwater and there’s nothing left?

Harvard’s Institute of Politics has just released its latest poll of 18-29 year olds, and reports that Hillary Clinton has a “massive” lead over Donald Trump. Over at the New York Times, however, Yamiche Alcindor says the new poll shows that Clinton has “struggled” with millennials and “will have to convince many young people that they should trust her to grapple with some of the nation’s biggest issues.” Nancy LeTourneau is annoyed:

That is the power of narrative. Once you buy into the idea that Clinton is having trouble with millennials, it is almost impossible to break out of it. In the back of Alcindor’s mind, she has to do better than a 28 point lead to be successful with young people. Who knows how high that bar is?

I get the exasperation with this, but the problem is that both the IOP and Alcindor are right. Clinton leads Trump 49-21 percent in the IOP poll, which is indeed a massive lead. At the same time, 49 percent support is less than Democrats usually get from 20-somethings. Like it or not, Clinton is less popular with young voters than any Democrat in the past two decades except for Al Gore. Is this because of the Bernie effect? Because of Clinton herself? Because third-party candidates are getting more attention than usual? That’s hard to say. But whatever the reason, Clinton is underperforming with millennials.

Now, at this point her underperformance is fairly modest compared to anyone other than Barack Obama. And she still has a couple of weeks to make up ground. It’s fair to say that she’s a little behind the usual pace for Democrats, but it’s not fair to regurgitate the narrative from two or three months ago when she was struggling pretty hard with millennial disaffection. It may not make for a great story, but sometimes the truth is a little bit boring.

With 13 days left until the end of the campaign, Donald Trump seems to have all but given up. He’s mostly promoting his hotels these days and has stopped all big-dollar fundraising. In fact, he seems as if he’d be pretty happy if Republicans lost in an epic wave election, which might make his own loss seem less of a personal humiliation and more a party failure. Given all this, I suppose this means that Republicans are resigned to losing and are probably putting their heads together to figure out how they can work with Hillary Clinton over the next four years in order to accomplish at least—

Eh? What’s that, Ilya Shapiro?

The Senate Should Refuse To Confirm All Of Hillary Clinton’s Judicial Nominees

Um, OK. That’s clear enough. Gonna be tough on the federal judiciary, though. Don’t big businesses need the courts to stay fully staffed so they can continue suing each other over dumb patent infractions? Maybe not. But anyway, Shapiro is just one guy. This is probably not a common opinion, right?

1. Dems will kill filibuster if that’s stopping SCOTUS confirmation. 2. SCOTUS works fine with 8. Wd work better w 7 https://t.co/kH6UyOjMHh

— John McCormack (@McCormackJohn) October 26, 2016

OK, fine: two guys. But surely wiser heads in Congress will prevail?

Jason Chaffetz, the Utah congressman wrapping up his first term atop the powerful House Oversight Committee, unendorsed Donald Trump weeks ago. That freed him up to prepare for something else: spending years, come January, probing the record of a President Hillary Clinton.

“It’s a target-rich environment,” the Republican said in an interview in Salt Lake City’s suburbs. “Even before we get to Day One, we’ve got two years’ worth of material already lined up. She has four years of history at the State Department, and it ain’t good.”

Welp, it’s sure sounding like the Republican Party has learned nothing and forgotten nothing over the past eight years. If this is how things go, they’re planning to double down on total obstruction starting on Day One—or even before that for Chaffetz. Then in 2020 they’ll wonder yet again why they have such a hard time winning the presidency. I wonder if it will ever occur to them that getting nothing done just isn’t a winning argument for a majority of Americans?

I’ve mentioned before that Obamacare premiums started out too low in their first year, which explains (a) why so many insurers have had trouble making money in the exchanges, and (b) why premiums increased so much this year. But maybe a chart will make this clearer.

This is based on data from Health Affairs last year, updated with the big increase in premiums this year. What it shows fairly clearly is that the cost of individual premiums dropped in 2014 when the Obamacare exchanges started up—even though Obamacare policies generally provided better coverage. When you factor in the big increase for next year, average premiums will have risen from $4,500 to $5,600 since 2013.

That’s an annual increase of 6.1 percent, about the same as the average annual increase in employer plans over the past decade.

The usual caveats apply. These are averages: some people do better, some do worse. And for people who qualify for Obamacare subsidies, the actual increase in the amount they have to pay is very small. Overall, though, the point here is clear: if premiums had just risen at a steady 6 percent per year, nobody would be bent out of shape. The reason this is hitting so hard is because insurance companies screwed up their projections when Obamacare started up and now they have to make up for it.

Justin Timberlake snapped a selfie in the voting booth yesterday, and lots of people were outraged that apparently there are laws against this. What happened to free speech!?!

Just for the record, then, there is a reason for selfie bans in voting booths: it prevents vote buying. After all, the only way it makes sense to pay people for their votes is if you have proof that they voted the way you told them to. Back in the day that was no problem, but ever since secret ballots became the norm vote buying has died out. Selfies change all that. If I give you ten bucks to vote for my favorite candidate for mayor, I can withhold payment until you show me a selfie proving that you voted for my guy.

How big a deal is this? I don’t know. Maybe we should go ahead and allow voting booth selfies. But the ban isn’t just a dumb bureaucratic rule. It’s a sensible attempt to prevent voter fraud that has very little cost.

In the same category of cluelessness as “Keep the government out of my Medicare,” one of my favorite dumb whinges comes from people who complain that the mainstream media isn’t covering something—and then illustrate that “something” by linking to a piece in the New York Times. Stephen Moore, in a brave attempt to keep his title of Stupidest Man Alive in the era of Trump, provides us with this classic of the genre today:

@StephenMoore on @Varneyco holds up today’s @washingtonpost,laments that there are no stories about Obamacare premium increases@ChuckLane1

— Matthew A. Duda (@nnw59) October 26, 2016

Every news outlet in America had a big headline about the Obamacare premium increases. It was plastered everywhere and blathered about endlessly on cable news. You could stay unaware of this only by hiding in a nearby fallout shelter like Kimmy Schmidt and not coming out for a week. As for the Washington Post, well:

In fairness, this is yesterday’s edition. The Post actually covered it before anyone else. I guess Moore somehow missed it.

Data! You want data! Sure, Obamacare premiums are going up and so are the subsidies. But how much are the subsidies going up? The chart below—which I want everyone to look at because it was a pain in the ass to create—shows this for the 15 states with the highest premium increases:

As you can see, subsidies are increasing more than premiums in every state—and by quite a bit. This comparison data is for a 27-year-old with an income of $25,000, and comes from Tables 6 and 12 here. (Arizona is literally off the chart: premiums increased 116 percent and subsidies increased 428 percent.) Here’s the same chart for the 15 states with the smallest premium increases:

There are plenty of caveats here. Premiums and subsidies will be different for different kinds of households. Upper middle-class families don’t get any subsidies at all. And this doesn’t tell us what the average net increase is, once subsidies are accounted for.

However, it gives us a pretty good idea that for a substantial majority of Obamacare users, the net amount they pay for health insurance in 2017 isn’t going to be much more than it was this year. For many, in fact, it will be the same. For those who shop around, it’s quite likely to be less.

Bottom line: if your income is low enough to qualify for a subsidy, there’s no need to panic over the Obamacare premium news. The higher premiums will help stabilize the market, and the cost will be covered almost entirely by Uncle Sam. Your pocketbook is safe.

This is from a guy who works for a healthcare advocacy group in New Mexico:

Just got off the phone w/ a consumer who was crying bc she couldn’t afford a 25% increase. With subsidies, her premium went down 1%. https://t.co/KYR4uhW6tC

— Colin Baillio (@colinb1123) October 25, 2016

Getting panicked calls all day about premium hikes. Every person I talked to was shielded by subsidies or on employer plan. #headlinesmatter

— Colin Baillio (@colinb1123) October 25, 2016

I don’t want to minimize the pain that this year’s premium hikes are going to cause for a subset of insurance buyers. But the vast majority of low-to-mid-income Obamacare users are eligible for federal subsidies—and as premiums go up, so do their subsidies. They may end up paying a bit more in 2017 for their health coverage, but probably no more than a few percent.

So yes: headlines matter. Or, at the very least, the first few paragraphs of news stories matter. Coverage of this issue should make it clear that the average price people pay will go up much less than 25 percent, and for low-income folks it probably won’t go up at all.

Last week, a self-driving truck delivered 50,000 cans of Budweiser from Loveland to Colorado Springs. This was obviously meant as a big FU to Coors, since the route “coincidentally” took all this frosty Bud right past Coors headquarters in Golden, Colorado. Most people, however, are interpreting this event as merely technological: it represents the dawn of the era of self-driving trucks. Tim Lee comments:

According to Otto’s blog post on the trip, “our professional driver was out of the driver’s seat for the entire 120-mile journey down I-25, monitoring the self-driving system from the sleeper berth in the back.”

But this doesn’t mean the nation’s truck drivers need to start working on their résumés. Technology like this may eventually displace human truck drivers, but the tech is several years away from causing mass unemployment. The key reason is that Otto’s self-driving technology is initially limited to highways. When the truck reaches ordinary city streets, it hands control over to a human driver to handle tricky traffic situations. This means that even after a truck is outfitted with Otto’s self-driving technology, it will still need a human driver in the truck.

Hmmm. “Several years” sounds ominously near-term, so truck drivers might want to start worrying about their jobs right now. Beyond that, there’s a way this could put truckers out of business well before that. Here’s how.

Pick a route that has a lot of truck traffic. Let’s say, Chicago to Cleveland. Outside of each city, you build a big truck depot and dispatch center. In Chicago, teamsters drive the trucks from the city out to the depot. Autopilots drive the trucks to the Cleveland depot, where a driver gets in and takes the truck to its destination. Rinse and repeat. The job of a truck driver is to drive back and forth from destinations in the city out to the depot, which they can do five or six times a day. Trucking firms save a ton of money even though the autopilot is designed for highway driving only.

Building the depots would be cheap and easy, since you don’t really need much there. It’s basically just a dispatch center. You could pretty easily have hundreds of them dotted across the country near all of our biggest cities. The only thing that would stop this from happening is the knowledge that they’ll only last a few years before they’re put out of business by fully automated trucks that can go from dock to dock with no human intervention. Either way, truck drivers are in big trouble.

There’s been a lot of talk about “fixing” Obamacare lately. Here’s my two-step plan:

1. Increase the subsidy levels.
2. Increase the penalty for not buying insurance.

That would pretty much do it. I could add lots of other small-bore things that need some tweaking, but why bother? These two things would do most of the job—and Republicans will never agree to them. They won’t agree to any of the small-bore stuff either. So take your pick. You can support a detailed 11-point plan for Obamacare that will never get passed, or you can support my 11-word plan that will also never get passed.

But since we’re all lightweight wonks around here, we should take a guess at how much we need to change the subsidy and penalty levels to make everything work. Basically, Obamacare’s big problem is that not enough young people are ponying up for insurance. To fix this, we need to get to a point where it’s cheaper for young people to buy insurance than it is to pay the penalty. This can be done by either increasing subsidies or increasing the penalty. Here’s my swag at what it would take:

You could increase subsidies by 100 percent and leave the penalty alone, or you could increase the penalty 250 percent and leave the subsidies alone. Or you can pick any point in between.

In reality, you could probably get by with smaller numbers, since nearly everyone will sign up if the penalty is within shouting distance of the net premium cost. You don’t have to literally make the penalty as high as the premium cost. I also assumed silver coverage in this chart, and you can assume lower numbers if you’re happy with kids buying bronze coverage.¹

Anyway, that’s it. This chart is my proposed Obamacare reform. It represents something of an upper bound, and I imagine that someone who has actual working knowledge of all this stuff could do a lot better. Call your congressman today and demand that this chart be made into law.

¹I’m not, especially, which is why I went with silver.