Kevin Drum

Let's Experiment With Universal Preschool

| Thu Sep. 24, 2015 2:43 PM EDT

I'm a considerable fan of early childhood education. Megan McArdle says she's tentatively in favor too, but "I am opposed to blind boosterism of such programs, the kind that confidently predicts marvelous results from thin empirical evidence, and briskly proceeds to demand huge sums be spent accordingly." I'm tempted to say this is a straw-man argument, but maybe not. There are a lot of cheerleaders out there. In any case, she offers a useful corrective for anyone who thinks the evidence in favor of universal preschool is open and shut. So what should we do?

I would like to see us experiment more with these programs. But the key word here is “experiment.” Which is to say we should: Try more programs....Take the programs that seem to work and scale them up to a larger group....Rinse and repeat [until we figure out what, if anything, works.] That would be the sane, sensible way to go about constructing policy in an important area.

But politically, how insane! Voters don’t want to hear about a decade or two of carefully planned research to help shape solid policy choices; they want to hear promises of immediate solutions to an immediate problem. That’s not a great way to make policy. But it’s a pretty good way to get elected.

I don't think these are mutually exclusive options. The 1988 Family Support Act might be a useful model here. Following a series of welfare reform experiments in the early 80s, it authorized additional research on a larger scale. Why not do the same thing with preschool? Offer substantial funding to states willing to participate in rigorous testing of preschool programs, with the goal of producing useful results in six or seven years.

This could be a substantial program, not just a few small-scale tests, which would certainly count toward any campaign promises made about universal pre-K. And the money would go to the states most eager to participate, which would be politically savvy. At the same time, it wouldn't cost as much as a nationwide program, which would make it easier to get through Congress. And finally, the promise of larger-scale testing would satisfy the demands of social scientists, who rightly point out that small-scale experiments don't always scale successfully into bigger programs.

I'm tempted to say that if Democrats and Republicans could agree on this approach for testing welfare reform in 1988, they should be able to agree on doing the same thing for preschool in 2017. That's not necessarily true, of course. Still, it seems like this kind of program would, at a minimum, be more likely to pass a divided Congress than full-blown universal pre-K legislation. Why not give it a try?

Advertise on

Why Is No One Talking About the Menace of the Pacific Ocean?

| Thu Sep. 24, 2015 1:36 PM EDT

Look, if we're going to have a wall on the Southern border and the Northern border, then I want a wall along the Western border too. I won't feel safe until we build one.

Shame about the view, but national security comes first.

Volkswagen's Emissions Conspiracy May Have Killed at Least 4,000 People Worldwide

| Thu Sep. 24, 2015 1:25 PM EDT

How many people did VW's NOx defeat device kill? Over the weekend I did a rough estimate and figured that over the past six years VW's excess NOx emissions probably killed about a dozen people in Southern California. Since then I've slightly revised my spreadsheet to account for an error, which increases my estimate to about 17 people killed. My figuring was based on:

  • 50,000 cars sold in Southern California between 2009-2014
  • 3,800 excess tons of NOx over six years
  • 0.0044 deaths per ton of NOx

VW sold 500,000 altered cars in the US and 11 million cars worldwide, so this extrapolates to about 170 deaths in the United States and about 3,700 deaths worldwide.

The number of cars sold is a solid figure, and as near as I can tell the estimate of 0.0044 deaths per ton of NOx is reasonable (this paper estimates a range of .0019 to .0095). But others have come up with higher mortality estimates than mine based on a much higher estimate of excess NOx emissions. So here are my calculations:

  • The ICCT, which discovered the violation, says VW cars "exceeded the US-EPA Tier2-Bin5 (at full useful life) standard" by 10-35 times depending on model.
  • The Tier2-Bin5 standard is 0.07 grams per mile.
  • If VW cars averaged 30x the standard, that's 2.1 grams per mile.
  • Based on (a) increasing sales year over year and (b) the fact that older cars have driven more miles, I figure that the affected cars have been driven about 1.6 billion total miles over six years.
  • That comes to 3.5 billion grams of NOx, or about 3,800 tons.

This extrapolates to 38,000 tons for the United States. That's over six years. But using the same excess emission rate of 30x that I did, the Guardian figures about 31,000 tons per year. That's five times my estimate.

My full spreadsheet is here. I invite comments.

Quote of the Day: "Carly Cut His Balls Off"

| Thu Sep. 24, 2015 11:41 AM EDT

It's been obvious to me for a while that the best way to get under Donald Trump's skin is to attack him where it really hurts. Don't call him a clown or an entertainer. That's water off a duck. But he genuinely cares about his reputation as a dealmaker. Hit that. Or his reputation for being tough. Hit that. National Review's Rich Lowry finally took this approach last night, and it worked:

"Let's be honest: Carly cut his balls off with the precision of a surgeon — and he knows it," Lowry said on "The Kelly File." Host Megyn Kelly was shocked. "You can't say that!" she said, before covering her eyes with a hand. "You can't say that."

....Trump quickly exploded on Twitter and wrote in a tweet: "Incompetent @RichLowry lost it tonight on @FoxNews. He should not be allowed on TV and the FCC should fine him!"

...."I love how Mr. Anti-PC now wants the FCC to fine me," Lowry tweeted, adding a hashtag: #pathetic....Lowry finally threw up a white flag and offered this tweeted compromise: "A deal for you, Donald: if you apologize to Carly for your boorish insult, I might stop noting how she cut your b**** off."

See? Easy peasy. Now I want someone to take on his dealmaking acumen. It shouldn't be too hard. That should really get him hot under the collar.

Quote of the Day: "He Didn't Think We'd Respond?"

| Wed Sep. 23, 2015 11:28 PM EDT

From a Carly Fiorina aide, defending their absurd smear today of Jeffrey Sonnenfeld:

After more than a decade of attacking Carly's character and misrepresenting her career, he didn't think we would respond? I guess he's not ready for the arena after all.

The context for this involves Trumpishly juvenile behavior from a whole cast of characters. Here's the nickel explanation. Back in 1997, Sonnenfeld was a professor at Emory University. That year, he accepted an offer from Georgia Tech to become dean of their business school. This apparently infuriated the Emory folks so much that they reverted to their kindergarten selves and hatched a scheme to tell Georgia Tech officials that Sonnenfeld had been caught defacing campus property. Georgia Tech promptly withdrew their offer, and Sonnenfeld was forced to resign from Emory.

This whole affair was basically ginned up out of whole cloth. A few years later, Emory withdrew their accusations of vandalism, and Sonnenfeld won multimillion settlements from both Emory and Georgia Tech.

But the fact that Sonnenfeld never did the things he was accused of didn't faze Fiorina. Last weekend Sonnenfeld wrote a Politico article criticizing Fiorina's tenure at HP, and that couldn't be allowed to go unanswered. In today's post-Trump Republican Party, the rule is, "He sends one of yours to the hospital, you send one of his to the morgue." So Fiorina had her aide send an email to reporters that said: "I'll give Sonnenfeld this: He would know something about getting fired. Of course, his was for vandalism of school property while he was at Emory."

Fiorina has obviously decided that imitation is the sincerest form of flattery and is bidding to become the new Donald Trump. If someone criticizes you, you hit him back in a comically heavy-handed way—and then hit him back again as a crybaby if he dares to object. This makes for good copy, and maybe that's all that matters. But Fiorina should watch out. Trump is such a clown that he somehow makes his attacks seem like the actions of a charming rogue, not a vicious hit man. When Fiorina does it, she just looks mean.

As near as I can tell, that's because she is. But it's still best not to let it show too much. Just ask Scott Walker.

If You Accuse Hillary Clinton of Lying, You Should Be Careful With the Truth Yourself

| Wed Sep. 23, 2015 8:09 PM EDT

I was noodling around this afternoon and decided to check out Drudge. Hmmm. A picture of Obama with 'horns' next to the pope. Some nutcases in California think the drought is part of a government weather-control conspiracy. China wants to control the internet. Obama has blocked a 13-year-old critic from following him on Twitter. Standard Drudge stuff. Then this: "FOURNIER: Come clean or get out..."

Ho hum. It was pretty obvious what the Fournier column was about, since he's been obsessed about Hillary's email server for months, but I went ahead and clicked anyway. I was pretty taken aback. He made three points at the top of the column:

  1. "The State Department confirmed that Clinton turned over her email only after Congress discovered that she had exclusively used a private email system."
    Nope. Fournier is referring to last night's Washington Post story, which says the State Department discovered it didn't know where Clinton's emails were. (Or Condi Rice's. Or Colin Powell's. Or Madeleine Albright's. Or much of anyone else's apparently.) Clinton turned over her emails when State asked for them.
  2. "A federal court has helped uncover more emails related to the Benghazi raid that were withheld from congressional investigators. Clinton has insisted she turned over all her work-related email and complied with congressional subpoenas. Again, she hasn’t been telling the truth."
    This is flatly false. The linked Politico story says nothing about Clinton not turning over all her work emails. It says only that the State Department has claimed executive privilege for a few documents—something with no relation at all to Hillary Clinton. From Politico: "The FOIA lawsuits provide a vehicle to force the agency to identify those emails, although the substance of the messages is not disclosed."
  3. "The FBI has recovered personal and work-related e-mails from her private server....The FBI has moved beyond whether U.S. secrets were involved to how and why. In the language of law enforcement, the FBI is investigating her motive."
    I guess this isn't flatly false, but "how and why" were words used by Bloomberg's reporter in the linked story. There didn't seem to be any special significance attached to them, and it's the airiest kind of speculation to say this means the FBI is investigating Clinton's motive. They've consistently said that she's not the subject of a criminal investigation. Why would they be investigating motive if they're not investigating any underlying crime?

That's three stories linked to, and all three were described in a badly misleading way. This is one of the reasons I usually pay so little attention to the Hillary email affair.1 It's been months now, and there's simply no evidence of anything other than unwise email practices and an unfortunate but instinctive defensiveness from Clinton over trivial matters. At some point, when nothing more comes up, it becomes clear that this is just the usual Clinton Derangement Syndrome at work. We passed that point a while ago.

Fournier has all but shouted that he's never trusted the Clintons and never will, and that's why he's so obsessive about this stuff. We all need a hobby, I guess. Still, he's a reporter. Deliberately distorting his descriptions of news accounts in the hope that no one will bother clicking on them is a bridge too far. He repeatedly claims that Hillary is lying, but Fournier is living in a glass house.

1Except today, I guess. But it's just an odd coincidence that this is my third post of the day on this "scandal."

Advertise on

It Might Be Time to Rethink How We Do Emissions Testing

| Wed Sep. 23, 2015 7:04 PM EDT

This sounds cool. Max Ehrenfreund writes today about Gary Bishop, a research engineer at the University of Denver, who has been working on real-life emissions testing for cars:

Bishop's laboratory has developed a roadside sensor, which he and his colleagues have been using for more than a decade to see how cars actually do on the street in several major cities....Authorities are now using the sensors in and around Denver and in a few other states as a supplement to conventional testing. The state sets up the sensors at highway on-ramps and elsewhere along the road. Drivers don't stop. They just roll between two rows of cones while a camera records the car's license plate and the equipment registers the emissions from the tailpipe, and go on their way. If a car produces at least two passing grades, the driver is spared the trip to the inspection station.

How about that. Can we get this in California, please? Of course, there's also this:

One of the cities where Bishop has worked is Tulsa, Okla., where emissions tests have never been required. The group has found that emissions from the cars in Tulsa are no worse than emissions in other cities where standards are enforced.

That's true. In a 2007 paper, Bishop concluded that emissions reductions have been about the same everywhere he's tested, regardless of whether periodic inspections are required. So maybe we need to ditch the big-government regulations that mandate the inspection regime altogether. Instead we could rely on spot checks of real-world emissions as a way of holding auto manufacturers accountable for complying with EPA standards, which suddenly seems like it might be the real problem after all. Let's get Jeb Bush on this.

It Sure Looks Like Hillary Clinton Didn't Have a Cunning Plan to Foil Congressional Investigators

| Wed Sep. 23, 2015 2:34 PM EDT

This happened yesterday while I was away from my desk:

The FBI has recovered personal and work-related e-mails from the private computer server used by Hillary Clinton during her time as secretary of state, according to a person familiar with the investigation.

The Federal Bureau of Investigation’s success at salvaging personal e-mails that Clinton said had been deleted raises the possibility that the Democratic presidential candidate’s correspondence eventually could become public. The disclosure of such e-mails would likely fan the controversy over Clinton’s use of a private e-mail system for official business.

Nobody seems to have made the most obvious observation about this: It pretty strongly suggests that Hillary Clinton was not trying to hide anything when she deleted personal emails from her server.

At the risk of boring my technically-minded readers, files on a computer work sort of like an old-fashioned card catalog in a library. If you "delete" a book by tearing up the index card, the book is still there. It might be harder to find, but with a little detective work you can still dig it up. Eventually, though, the book will truly disappear. Maybe someone steals it and no one cares. Or the library needs more space and gets rid of all the books with no index cards. Etc.

This is how computers work. When you delete a file, you're just deleting the index card. The file is still there on the hard drive. Eventually, though, the file will truly disappear. Maybe another program writes over the file. Or you run a disk defrag program and whole sections of the disk get written over. Etc. Some files will get permanently deleted within days. Others might stick around for years. It's just random chance.

Needless to say, things don't have to happen this way. If you want to make sure that a file is well and truly deleted, it's easy to do. Anyone with even a smidgen of computer experience either knows how or knows how to find out. Here's one way, which took me ten seconds to Google. If I were really serious, I'd take the time to read a bit more, and also make inquiries about backups. This is IT 101.

But apparently Hillary didn't ask about any of this stuff. No one on her staff brought it up. They just pushed the Delete key and the emails disappeared. The IT folks were never involved.

These are not the actions of a staff trying to stonewall FOIA requests or foil a congressional committee. Any bright teenager could have done better on that score. By all the evidence, Hillary is telling the truth. She just told her staff to delete personal emails and turn over the rest to the State Department. There was nothing more to it.

But no one's reporting it that way. Peculiar, isn't it?

Jeb Bush Has No Clue About Business Regulation

| Wed Sep. 23, 2015 1:34 PM EDT

Jeb Bush today in the Wall Street Journal:

To understand what is wrong with the regulatory culture of the U.S. under President Obama, consider this alarming statistic: Today, according to the World Bank—not exactly a right-wing think tank—the U.S. ranks 46th in the world in terms of ease of starting a business. That is unacceptable. Think what the U.S. could be and the prosperity we could have if we rolled back the overregulation that keeps us from ranking in the top 10.

My goodness. That does sound unacceptable. Still, it never hurts to check up on these presidential candidates, does it? So let's click the link.

Sure enough, the World Bank ranks the United States 46th in ease of starting a business. But there's an asterisk next to that. Let's scroll down and see what it says: "The rankings of economies with populations over 100 million are based on data for 2 cities." Hmmm. It turns out the World Bank is ranking the United States based on starting up a business in New York City. That seems to tip the scales a wee bit, no?

But let's soldier on. New Zealand ranks first in starting a new business, so let's see how they work their magic. Here's the World Bank's comparison:

So it takes half a day in New Zealand and four days in New York City. Really? Half a day to start up a new business? Maybe they're not using the same definition of "starting" that I am. Let's check out the details for New York City. Here they are:

Now I get it. This isn't about getting a business up and running. It's solely about registering a new business. And it's got nothing to do with any of Obama's regulations. It's all about state and local stuff. The only part that's federal is getting an EIN number, which is free and takes a few minutes. I'm not sure what Jeb Bush thinks he's going to do to streamline this.

Bottom line: This is completely meaningless. It's a measure only of how long it takes to register a business, and it's only for New York City. And even at that, it takes only four days and costs $750. This is not stifling American entrepreneurship.

But wait! There's more. The World Bank does have a broader "Ease of Doing Business" rank that takes into account the things you need to do to get up and running: construction permits, electricity, credit, paying taxes, enforcing contracts, etc. As it happens, the bulk of this stuff is still state and local, and has nothing to do with Obama or the federal government. Still, let's take a look since Jeb chose not to share it with us for some reason. Where does the United States rank on this measure?

The World Bank has us in seventh place. We're already in the top 10 that Jeb is aiming for. Mission accomplished!

POSTSCRIPT: Jeb has many other statistics in his piece, and I'd take them with the same grain of salt as his World Bank numbers. He also promises that in his administration every regulation "will have to satisfy a rigorous White House review process, including a cost-benefit analysis." Apparently he doesn't realize that this is already the case. As for the outrageous regulations he promises to repeal on Day One, this would mostly benefit big campaign donors, not the yeoman entrepreneurs he claims to be sticking up for. No big surprise there, I suppose.

For Blue-Collar Men, Life Looks Increasingly Dismal

| Wed Sep. 23, 2015 12:19 PM EDT

Here's a merger of two charts that have made the rounds recently. The first, from Brookings, shows a familiar pattern: the median pay of a man employed full-time has dropped substantially since 2010. The second, from the Kaiser Family Foundation, shows that health care deductibles have risen substantially since 2010.

Put them together and you get the chart on the right. The light red line is bad enough: blue-collar men earn about $3,000 less than they did five years ago. The dark red line is even worse: if you factor in rising deductibles, they're earning $3,500 less than they did five years ago.

This explains a lot of the discontent of the past five years, especially among working and middle-class white workers. In theory, health care is getting better every year, and if you take that into account then total compensation starts to look a little better. Technically, this is true. But think about it from the average worker's point of view:

  • His cash wages have gone down.
  • Health care may be getting better, but that's mostly invisible. It doesn't seem any different than usual.
  • But high deductibles provide an incentive not to see the doctor when something minor is bothering you. So, in practice, health care actually seems not merely the same as always, but actually a bit worse and a bit more of a hassle. Either you ignore the minor stuff or else you go in and, thanks to higher deductibles, end up paying an infuriatingly high bill.

For your average blue-collar man, here's what life seems like: wages are down, health care is more expensive, and you have to spend a lot more time worrying about whether it's worth it to see your doctor. There's not much to like in this picture.