The opioid crisis and the border Wall

The states hardest hit by the opioid crisis are a long way from the Mexican border. Trump’s Wall remains a solution in search of a problem.

One of the sillier talking points in the Wall debate is that we need a physical barrier to keep opioids from coming into the country from Mexico.  Various commenters have pointed out that: (1) The fentanyls, which are the fastest-growing segment of opioid use and overdose deaths, mostly come directly from China; and (2) What does come across the U.S./Mexico border comes through overwhelmingly by common carrier at ports of entry; it isn’t backpacked through the desert by immigrants.

A point I haven’t seen made, and didn’t know about until Kevin Drum posted this graph based on data from the CDC’s Morbidity and Mortality Weekly Report, is that the crisis isn’t concentrated anywhere near Mexico. All of the hardest-hit states in terms of opioid mortality rates are east of the Mississippi and north of the Tennessee, about as far as they could be from the Rio Grande. Of the four states that actually border Mexico, New Mexico and Arizona are in the middle of the pack, while California and Texas rank 45th and 47th.

So Trump’s Wall remains a solution in search of a problem.

Liveblogging the midterms, Part IV

I might not have to slit my throat after all.

The Senate is a lost cause, and the Georgia and (probably) Florida governorships have gone the wrong way, but a couple of good results in the House have turned the 538 model to a 2/3 chance of Democratic control, while the NYT reports >95% and predicts 231 Democratic seats (+34) based on a popular-vote margin of almost 9.5%.

That means two years of investigating Republican misconduct, and in particular Adam Schiff having Mueller’s back.

Now if we can just take out Kobach and Walker …

Kobach has now officially lost. That makes up for a lot.

And 538 is now calling it 90% likely that the Democrats take the House.

Jared Polis has been elected Governor of Colorado.

Spanberger beat Dave Brat in Virginia.


Liveblogging the Midterms, Part III

Well, Ollie, ain’t that a revoltin’ devlopment!

I’ll raise my hand now and admit that I was expecting the Democrats to outperform their polling. Apparently, no such luck. Looks as if racism was a good strategy for the GOP.

Donnelly is losing Indiana; both Florida races are too close to call; no surprising House pickups so far; a couple of promising candidates have lost. And if we had to win one possibly marginal Senate seat, did it have to be that crook Menendez?

538 now projects only a 60% chance that the Democrats will take the House, with an expected gain of 25 seats, and expects a net gain of two Senate seats for Republicans.

8:49: HAP–I suppose it’s a good thing that Menendez was reelected in NJ. A real embarrassment. I know it’s a good thing that Florida has voted to restore the voting rights of more than 1 million former felons.

Less than 100,000 vote margin in Fla gov race. Republicans well-positioned to win both Florida races, as Mike Grunwald warned months ago.

O’Rourke doing well in early voting, but it’s hard to know.

8:54. Republicans picked up Indiana Senate seat.

Democrats seem to be under-performing overall polling numbers. Not a great night so far.

8:58 MSNBC projecting that Democrats will 224+/-8, with 65% chance of winning majority.

8:59: Manchin held on….

9pm: Cruz has slight lead, but too early to call. Abbott held on in governor.
Wis governorship too close. Tammy Baldwin reelected.

9:04 Marsha Blackburn won in Tennessee. Gillum has a big hill to climb in Florida.

538 gives Democrats 56% chance of taking House. Looks like MS will go to a special election.

As depressing as tonight is, Democrats now at 57% probability of taking the House.

Striking that Democrats who ran as Democrats were sometimes disappointed. Democrats who ran as Republicans were also sometimes disappointed. McCaskill voted against Kavanaugh but also expressed anti-immigrant sentiment.

[We are adjourning to the top of the hour.]

Liveblogging the midterms, Part II

7:45 So far, so good. FiveThirty Eight started the night predicting that the Democrats would lose a Senate seat and pick up 35 House seats. That’s now a break-even in the Senate and +45 in the House.

Gillum and Nelson seem to be running about even with each other in Florida, about three points ahead of their opponents with more than half the vote in.

7:58 Now the 528 predictions are back down to R+1 in the Senate and D+41 in the House. Donnelly isn’t looking good in Indiana; he was even-money on PredictIt earlier in the day, but he’s now 4-to-1 against. Apparently the rural counties are coming in hard for his opponent.

First pick-up call of the night: Wexton over Comstock in VA-10. That was expected.

8:24 Doesn’t look good for Gillum right now. Wow. Florida looks like a nail-biter.

CNN continues to over-interpret early results without telling us where early votes are coming from.

Charlie Baker wins in Mass reminder that socially-moderate Republicans could be very competitive in blue states, if GOP primary electorates would nominate mainstream candidates

Wolf Blitzer yells a lot. If he were on Twitter, he would be all-caps. These early results would be so much more informative if they were embedded in some sort of predictive-analytic model based on prior state patterns.

8:25 No surprise: Pritzker and Hogan win governorships.

8:26 James Carville says this will not be a Democratic wave election. Seems a little dour.

Starting a new post now.

Liveblogging election returns

The Reality-Based Community
will be liveblogging tonight’s results.

Harold Pollack, Stuart Levine, Mike O’Hare, and I will be live-blogging the midterms. (We’ll be chatting by email, and I’ll be posting the juicy bits here.)

First comment from me: It’s time for Democrats to make vote suppression and elections integrity a campaign issue. The only people that will turn off are the hardcore Trumpist Republicans Democrats were going to lose anyway.

If the Democrats take the House, they can pass an appropriate bill (which the Senate will ignore) and then start tacking it to various pieces of “must-pass” legislation.


CNN is putting up some exit polling. Overwhelmingly, voters think the economy is in good shape (68-31) but the country is on the wrong track (41-56). That sounds right.

Trump’s job performance among exit-polled voters is underwater, 44-55. That’s pretty close to pre-election national polling.

Opinions of the Republican Party are net unfavorable (43-54, of the Democrats net favorable 50-46).

Generic ballot also has the Democrats 10 points up, also consistent with pre-election polls, or just a tad on the high side.

I’ve always found it bizarre that the networks report raw vote without reporting votes in previous elections from the same precincts, which would make the raw vote interpretable. That shouldn’t require much effort.

Predictably, is providing a contrast with CNN’s idiot chatter. Here’s how the thing is done, folks:

6:49 PM

Looking for signs of what’s happening with Democratic incumbent Sen. Joe Donnelly in Indiana? In Bartholomew County, which had 1 percent of the state’s vote in 2016, a bit more than one-third of its precincts have reported. Trump won it by 33 points in 2016, and Donnelly lost it by 8 points in 2012. Currently Donnelly trails there by 9 points, which is close to his 2012 result when he won statewide by 6 points.

If those 1/3 of the votes are representative of the county – a big if – Donnelly is in good shape; he’s only doing a point worse tonight than he did six years ago, when he won by six points. (Still not clear to me why it’s so hard to do straight precinct-by-precinct comparisons.)

Now here’s some cheerful news from Twitter:


On the Kim Davis race in Rowan County, with 12 of 19 precincts reporting:
Davis (R): 1783
Caudill (D): 2172
@heraldleader @BGPolitics

Closing this thread down and starting a new post.

How big a problem is heavy cannabis use?

Annie Lowrey has a well-thought-out piece about cannabis use disorder in The Atlantic. In my view, this is the most under-covered topic in the debate about whether – and, more importantly how – legalize cannabis.  I’m especially grateful to Lowrey for making me sound coherent; that doesn’t always happen.
Unfortunately, the editor who added the headline and subhead to the story misstated matters fairly significantly. (This happens so often that I now insist on having a voice in the headlines that run over my essays.)

Continue reading “How big a problem is heavy cannabis use?”

The verdict against Monsanto

A California jury has awarded $289 million to DeWayne Johnson, a groundskeeper who has non-Hodgkins lymphoma and claims his disease was caused by glyphosate, Monsanto’s blockbuster pesticide marketed most prominently as Roundup.

Michael Hiltzik of the LA Times reviews the facts of the case and the scientific dispute surrounding glyphosate’s role, or lack thereof, in causing cancer. He points out, reasonably, that a courtroom is a lousy places to resolve scientific disputes; how many of the twelve jurors could define “statistical significance” or “type II error”? And the notion that non-experts can pick out the truth in a swearing contest between experts is simply laughable.

Hilzik notes that alternatives, including “bringing these cases before specialized tribunals or setting up public funds for victims of certain products,” “all have their own flaws,” and concludes with an expert’s view that the jury trial “is a highly imperfect process,” but that “like democracy, it’s the best we have.”

I doubt it. Continue reading “The verdict against Monsanto”

Justice Department releases FISA warrant applications on Carter Page

My quick analysis:
400-something pages, mostly redactions, and the rest mostly boilerplate that gets repeated from application to application.
Still, what’s left is interesting. And, naturally, the documents make complete nonsense of the conspiracy theory Devin Nunes and his House Intelligence Committee Republican colleagues have been pushing.
Everything about the Steele Dossier – including Steele’s decision to talk to the press just before the election – was fully revealed to the court, and there was plenty of non-Dossier support for the idea that Page was acting as a Russian agent. Moreover, the extension applications continue to recite that the Bureau believes “Source 1’s” (that is, Steele’s) “reporting herein to be credible.” If the wiretaps conducted under the warrant had in any way disconfirmed Steele’s material, the Bureau could hardly continue to recite that Steele’s reporting was credible.
First application in October 2016, extended January, April, July. (90 days is the limit for a FISA warrant; an extension requires a new application.
Each application is signed by the FBI Director and the Attorney General (or substitute after the Sessions recusal). October and January applications are signed by Sally Yates as AG.
Last two are signed by Boente (April) and Rosenstein (July). Comey signs as FBI Director the first three times; Wray signs in July.
[Footnote: I was close to the parallel process for wiretap applications, requiring sign-off by an Assistant Attorney General. That was taken enormously seriously, the signature was not a rubber stamp. Each application was read in detail by someone on the AAG’s personal staff, and more than one application was sent back or refused outright. Hard to believe FISA applications aren’t taken comparably seriously.]
Presumably much of the redaction is about the product; every extension has to show that the previous 90 days were productive. The Times counted pages: 66 pages  in the original, while the extensions counted 79 pages, 91 pages and 101 pages, suggesting that there was significant product. But that was already clear from the fact that the extensions were requested and granted. Courts frown on continuing to drill dry holes.
Basis of the first application was the FBI belief that Page was “collaborating and conspiring with the Russian Government” and that “the Russian Government’s efforts [to mess with the campaign] were coordinated with Page and perhaps other individuals associated with Candidate 1 [Trump]
Can you say “No collusion”? I was sure that you could.
Update: Leah McElrath points out that this assertion – like the assertion of the reliability of Steele’s reporting – is repeated verbatim in the three extension applications, which it couldn’t be if the wiretaps had failed to confirm it. More detail from Twitter account @PwnAllTheThings.
The application recites that Carter was a knowing intelligence agent, recruited by three named SVR officers acting under Non-Official Cover, one of whom, Buryakov, was arrested in January 2015 and pleaded guilty to a violation of the Foreign Agents Registration Act (FARA) by acting as an unregistered foreign agent in May 2016, getting 30 months.
Page’s mission is said to have been “clandestine intelligence activities (other than intelligence gathering activities).” If that applied to Buryakov, that might explain why he was charged with a FARA violation rather than the more serious charge of espionage.
Comic relief: In February, 2017, Page asks the Voting Rights Section of the Civil Rights Division to investigate whether the Clinton campaign had engaged in “severe election fraud”  involving “disinformation, suppression of dissent, hate crimes, and other extensive abuses” by saying mean things about Page.
Conclusion: The warrant was issued on the basis of the FBI’s belief that Carter Page, a Trump adviser, was knowingly working for the Russians, and that other Trump campaign personnel might be doing the same. It was then extended three times, strongly suggesting that the taps yielded, and continued to yield, valuable counterintelligence. And the terms of those extension applications strongly suggest that the Steele Dossier, and the claim that Page was conspiring with Russia to help Trump, kept looking good.
It gets harder and harder to credit the good faith of anyone who still insists that there is doubt that Russia, as a matter of national policy, interfered with the 2016 election to secure victory for its favored candidate, and that at least one Trump campaign official knowingly helped.

Does the Constitution authorize immigration restriction?

Continue reading “Does the Constitution authorize immigration restriction?”

How to Have Less Crime and Less Punishment

This is the talk I gave at NYU’s Hoffinger Colloquium this spring.

Everything I’ve learned about crime over 40 years, in 60 minutes.

It’s a mix of theoretical principles and practical suggestions.

“We massively over-punish
and haven’t gotten much in the way of results.
That doesn’t prove that punishment doesn’t work.
It just means we’re doing it wrong.”

Talk starts about 4 minutes in.