Month: March 2021

Here's a short list of questions that will determine if you are a person who "follows the science."

Don't overthink this. And don't get pedantic. I know all about genetic drift and deadweight loss and the Cutter incident. But anyone with a decent sense of what science tells us should be fine agreeing with this list without insisting on a bunch of caveats.

1. The theory of evolution by natural selection is correct.
2. Cognitive abilities in human beings are significantly—but not entirely—controlled by biology.
3. Man-made climate change is very real and very serious.
4. Tax cuts do not pay for themselves.
5. Astrology is nonsense. Generally harmless, but still nonsense.
6. Childhood vaccines are perfectly safe, and in particular they don't cause autism.
7. GMO-based food is safe to eat.
8. You are personally committed to judging research papers by consistent standards of sample sizes, proper controls, well-defined hypotheses, reasonable modeling, and general overall rigor regardless of the conclusions they draw.

If you answered YES to all these items, your score is 1. If you did anything else your score is 0.

I would like to emphasize that if your score is 0, that doesn't mean you're a bad person. I have lots of perfectly decent friends in this category. All it means is that you aren't fully dedicated to following the science regardless of where it goes.

This is a Schmidt's red-tailed monkey at the San Diego Zoo. He looks so sad, doesn't he? But I'm sure he has no idea that the set of his face happens to coincide with the human conception of a sad look.

Then again, maybe if someone figured out his favorite food he'd suddenly look a lot happier. You never know.

A Twitter conversation earlier this morning got me interested in looking at some basic data on remote work in America. The starting point came from Adam Ozimek of Upwork, who surveyed corporate executives and concluded that they were much more open to the idea of remote work than they had been before the pandemic. After all, remote work has lots of advantages.

Likewise, there have been a lot of news reports about people moving during the pandemic, usually from crowded cities to smaller towns where the risk of being infected by COVID-19 is perceived to be lower. After all, if they can continue doing their jobs from anywhere, why not?

Let's take a look at a couple of pieces of data. First off, here is geographic mobility over the past few decades:

The data for 2020 is estimated from US Postal Service change-of-address cards, and the entire chart is only for (a) permanent movers who (b) move across county lines. As you can see, there's been a long-term decline in permanent moves, and even the pandemic didn't change that by more than a hair.

Here is the pre-pandemic trend in working at home, also based on Census data:

This chart is interesting but inconclusive. One interpretation is that despite the fact that remote work has lots of advantages, it's only grown by a meager two percentage points over the course of 20 years. That doesn't suggest an awful lot of enthusiasm for it even though—contrary to much punditry—I can assure you that corporate execs are keenly aware of its pros and cons and have been pondering it closely for years.

A different interpretation is that at 5.5% there's a ton of room for growth since, roughly speaking, about a third of Americans work in jobs that could be done at home. All it will take is for hiring managers and employees to be exposed to it and decide they like it.

These raw data posts aren't generally a place for me to make an argument. They're just places to present interesting data. So I'll just say that this is a good area to keep an eye on to test my basic hypothesis about pandemic behavior: It will very quickly go away and corporate life will return to its 2019 baseline. People don't really want to move, even as the cost of big cities continues to increase, and they've never warmed to remote work despite its myriad advantages. The big question is whether merely being forcibly exposed to it will change this dynamic significantly even though the low-hanging fruit has already been plucked. I don't think it will. Adam disagrees. Come back in 2023 and we'll settle this once and for all.

Republicans have introduced a huge raft of bills in red states that would restrict voting in federal elections. And make no mistake: these bills are partisan cruise missiles aimed solely at reducing the Democratic vote. They are loathsome.

But they are also pretty conventional. For the most part, they include all the old-school Republican favorites like restrictions on Sunday voting, photo ID laws, and so forth. These are all things Republicans have been promoting for years, and none of them have turned out to be all that effective. In some cases this is because they simply don't affect very many votes, while in others it's because they spur a backlash that prompts Democratic-leaning voters to turn out in higher numbers.

The big new addition to this arsenal is an attack on mail voting. Republicans are apparently convinced that this cost them the 2020 election, but it turns out these bills are just more of the same old ineffectual rubbish. Ian Millhiser describes a natural study of Texas voters:

Texas is one of a handful of states that discriminate on the basis of age when determining who may vote absentee: Voters over the age of 65 are allowed to request an absentee ballot, but most voters under that age are required to vote in person. The researchers at Stanford’s Democracy and Polarization Lab compared turnout rates among 65-year-olds in Texas’s 2020 elections — that is, among voters who were eligible to vote absentee — to the turnout rates among 64-year-olds who mostly could not vote by mail.

They found that “65-year-olds in Texas turned out in 2020 at almost exactly the same rate as 64-year-olds, even though roughly 18% of 65-year-olds voted absentee while only 3% of 64-year-olds voted absentee.”

The end result was that 65-year-olds were more Democratic than 64-year-olds by 0.2 percentage points, a difference too small to even be meaningful.

Other studies have shown the same thing: All of these anti-mail measures have only a very tiny effect. It may make Republicans feel good to demonstrate a whirlwind of activity that Fox News is happy to spin as "protecting against fraudulent votes," but in the end it doesn't really buy them much.

On a moral plane, this does nothing to excuse what they're doing. But on a practical plane, it means they're mostly wasting their time.

Here’s the officially reported coronavirus death toll through March 8. The raw data from Johns Hopkins is here.

The Un-Carrier is apparently becoming less and less Un now that it's absorbed Sprint and no longer needs to act like a hungry underdog:

T-Mobile will automatically enroll its phone subscribers in an advertising program informed by their online activity, testing businesses’ appetite for information that other companies have restricted.

....A T-Mobile spokeswoman said the changes give subscribers advertising that aligns with their interests. “We’ve heard many say they prefer more relevant ads so we’re defaulting to this setting,” she said.

Uh huh. They're exhausted from fielding constant requests to please share app usage with anyone who wants it, so they're finally caving in to what today's consumers have loudly said they want. Either that or they're responding to the fact that they're now one of the Big Three and don't have to compete as hard as they used to. This means they can afford to try out some profit-friendly tweaks and see if it pisses off very many people.

In any case, you can always opt out from this. Naturally. And as the Un-Carrier, they surely make this easy. So, since I'm a T-Mobile subscriber, I decided to opt out. Here's how it went:

1. Go to T-Mobile site.
2. Log in to account.
3. Dig up password since it's been years since I last used it.
4. Stop! Password too old!
5. That's OK, I'll keep it.
6. Sending a 6-digit code to your phone. Please enter it below.
7. Ding. OK, got it.
8. Now change your password anyway.
9. What?!? OK fine. Here's a new password.
10. Account page pops up.
11. Browse around for something that looks like it might be an opt-out.
    
12. Hmmm. Change my plan. Add a person. Report stolen device. Manage add-ons. See latest deals. None of those seem very helpful. Keep looking.
13. Ah. Up in the corner: Edit Profile Settings. Give it a try.
14. Billing. Caller ID. Scroll. Language settings. Blocking. Scroll some more.
15. Aha. At the very bottom, Privacy and Notifications. Click.
16. Marketing Communications Preferences? Maybe. But no.
17. Notifications? No.
18. Last choice: Advertising and Analytics. Not sure what that is, but . . .
19. Hooray! Turn off both "Use my data for analytics and reporting" and "Use my data to make ads more relevant to me." After all, who knows which one is really the opt-out?

All done. Wasn't that easy? Of course, T-Mobile's 80 million customers will never be informed of this change in the first place,¹ so most of them will be entirely spared the cognitive discomfort of deciding if they want to opt out. And for the few who do somehow learn about it and do want to opt out, they should have no trouble figuring out the EZ steps above. So why do douchenozzles like me keep making such a big deal out of this stuff?

¹Outside the confines of some new version of the T-Mobile privacy policy, that is.

Should we just give poor people more money? Over at Vox, Sigal Samuel summarizes a recent experiment:

The city of Stockton, California, embarked on a bold experiment two years ago: It decided to distribute $500 a month to 125 people for 24 months — with no strings attached and no work requirements. The people were randomly chosen from neighborhoods at or below the city’s median household income, and they were free to spend the money any way they liked.

....The most eye-popping finding is that the people who received the cash managed to secure full-time jobs at more than twice the rate of people in a control group, who did not receive cash. Within a year, the proportion of cash recipients who had full-time jobs jumped from 28 percent to 40 percent. The control group saw only a 5 percent jump over the same period.

I suppose I've read a hundred pieces like this over the past few years. Some describe American experiments. Finland did one a few years ago. Africa is a popular place for philanthropies to run UBI projects. The problem is that none of them really tests the Universal Basic Income thesis in a way that matters for the United States. This would have the following requirements:

• The test takes place in the US. Even a well-designed project simply doesn't tell us much if takes place in, say, Nigeria, with its wildly different culture and enormously lower living standards.
• It needs to include enough people to provide the power needed to draw serious conclusions. At a guess, that's a bare minimum of 100 people, and probably more like 500 or so.
• The amount of money needs to be substantial enough to make a real lifestyle difference. Ideally, it's enough to allow someone to quit working entirely if they're willing to live on a low income. Here the bare minimum is probably $1,000 per month.
• The project includes a control group that gets at least a small amount of money (so that the results aren't skewed too badly by the mere act of receiving cash).
• Needless to say, the test group and the control group can't know that the other exists.
• The test has to be very long term. Probably ten years minimum. If it's shorter than that, the test subjects know that they're just getting a short-term infusion of cash. They will treat it very differently from knowing that they are guaranteed a significant flow of strings-free cash for the rest of their life.

Now, there's an obvious problem with my requirements: The only kind of test that qualifies is very expensive and takes a very long time to return results. Including both the cash and the people to run the experiment, we're probably talking $30-100 million depending on how rigorous the test is.

This is hardly impossible if some billionaire gets really interested in the idea, but that's what it would take. In the meantime, the raft of experiments that are done in poor countries; or last for only a year or two; or involve small monthly stipends—well, they just don't tell us much. I honestly don't think there's an answer to this.

My latest cancer screening is boringly predictable:

For the past ten months my M-protein level has been between 0.35 and 0.40. This is not the lowest it's ever been, but it's the steadiest it's ever been. And while it would always be nice to get lower readings, this is basically good news and I hope it stays this way for a nice long time.

Let us greet the new week with a glorious sunrise on the Blue Ridge Parkway near Blowing Rock, North Carolina. I think this matches my feelings at the moment about the pandemic: a glimmer of hope that daylight is finally returning, but still a while to go before the darkness recedes and we no longer need to keep the lamps on.

In a post that happens to include a bit of mulling over the fate of Western civilization, Jay Nordlinger adds this aside:

(Want to know some good news? The threat of radical political Islam receded faster than many of us expected. It still lurks, of course — what doesn’t? But I well remember the concerns of the first decade of this century. Many of us were settling in for a long twilight struggle. In any event . . .)

This reminds me of something to brag about. Several years ago I predicted that the region from north Africa to the Mideast to central Asia would soon see a substantial decline as a source of terrorism. The reason was simple:

Unlike the US and Europe, this region didn't begin phasing out leaded gasoline in the '70s and '80s. That had to wait until the late '90s for some countries and a few years later for others.

Significantly, both Egypt and Saudi Arabia, ground zero for the production of terrorists, phased out lead around the year 2000, which means that teens and young men in both countries have now largely grown up without lead poisoning—and within a decade everyone under the age of 30 will be lead free. In the same way that this led to a plummeting crime rate in the US, it was almost certain to lead to a plummeting terrorism rate in these two countries.

Other countries have lagged behind a bit, but most of them phased out leaded gasoline in the mid-2000s or a few years later. The laggards are Algeria, Iraq, and Yemen, and it's no coincidence that all three remain very violent places.

I hardly need to say that human history is full of violence and revolution that had nothing to do with lead poisoning. Nor is lead the sole reason for the decline of Mideast terrorism. Nevertheless, the lead epidemic of the late 20th century had a huge impact, and as it fades away the level of senseless violence—on average—is likely to decline as well. That's true all over the world, not just on the police blotters of the United States.