The Wall Street Journal's Devlin Barrett has the latest from his disgruntled sources at the FBI who are hell-bent on proving that the Clinton Foundation is corrupt:

Much of the skepticism toward the case came from how it started—with the publication of a book suggesting possible financial misconduct and self-dealing surrounding the Clinton charity. The author of that book, Peter Schweizer—a former speechwriting consultant for President George W. Bush—was interviewed multiple times by FBI agents, people familiar with the matter said.

…In February, a meeting was held in Washington among FBI officials, public-integrity prosecutors and Leslie Caldwell, the head of the Justice Department’s criminal division…Following the February meeting, officials at Justice Department headquarters sent a message to all the offices involved to "stand down," a person familiar with the matter said.

…The FBI had secretly recorded conversations of a suspect in a public-corruption case talking about alleged deals the Clintons made, these people said. The agents listening to the recordings couldn’t tell from the conversations if what the suspect was describing was accurate, but it was, they thought, worth checking out.

Prosecutors thought the talk was hearsay and a weak basis to warrant aggressive tactics, like presenting evidence to a grand jury, because the person who was secretly recorded wasn’t inside the Clinton Foundation.

Holy crap. The field agents got started on this because they read Clinton Cash, the latest in a 25-year stream of books insisting that the Clintons are our era's Borgia family. Then they got even more interested because some guy—literally just some guy—was recorded blathering about the foundation.

This is what Hillary Clinton has faced ever since 1992. It started in Arkansas, where this kind of character assassination is basically just political sport, and then got exported to Washington, DC, when Bill Clinton ran for the presidency. There's nothing in Schweizer's book except yet another CDS victim1 insisting that if two things happened at the same time, somehow the Clintons were responsible for it and he's got the proof right here!

And now it turns out the FBI is full of middle-aged white guys who apparently read the book, listen to lots of Rush Limbaugh, and just know that if they look a little harder they'll find the one Jenga brick that causes the whole Clinton edifice to finally collapse. Jesus Christ.

1Clinton Derangement Syndrome.

This chart is a follow-up to my post last night about response rates in polls. It's from the paper that started the whole thing, published earlier this year by Andrew Gelman and three other researchers. They analyzed the 2012 campaign, and what they found was that polls were far more variable than actual voting intentions. The red line is what the polls looked like in real time. The black line is what they look like when you control for different response rates:

The first dotted line is the first debate. Remember how Obama did so poorly and plummeted in the polls? It turns out he didn't, really. Obama fans just stopped responding to polls, producing the illusion of a 10-point collapse. In reality, he only dropped about 4 points. In fact, during the last six weeks of the campaign, Obama's support was never more than a couple of points away from 52 percent.

Moral of the story: even poll averages bounce up and down misleadingly. In reality, there's probably no more than two or three points of change in actual voting intentions during the last month of the campaign. And in the last week? Practically none at all.

A few days ago the New York Times ran a big front-page story about GMO crops. Roughly speaking, there were two takeaways: GMO crops don't improve yields and don't cut down on pesticide use. So why bother?

That sounded interesting, but I decided not to write about it. The story was pretty shallow in its use of statistics. It assumed that you can compare different countries without controlling for anything (different soils, climates, crops, etc.). And it seemed to suggest that American farmers must be idiots, because they keep buying GMO seeds even though they're worthless. Put together, this was a little too much for my spidey sense, and it would have taken a long time to educate myself enough to really figure out what was going on. So I passed.

Yesterday, however, Nathanael Johnson of Grist took a close look at the Times piece and concluded that it was pretty misleading. You can read Johnson's piece if you want all the gory details, but it contained a pretty interesting example of how to mislead with statistics, which I figured I should pass along. The Times story says that in the US, despite the wide use of GMO seeds, use of insecticides has fallen but the use of herbicides has gone up. In France, by contrast, where they don't use GMO seeds, "use of insecticides and fungicides has fallen by a far greater percentage — 65 percent — and herbicide use has decreased as well, by 36 percent." Using data from Andrew Kniss, here's what this looks like in chart form:

It does indeed look like France is doing pretty well. But what if you look at the raw numbers, rather than percentages? Here's a chart from Kniss:

In the case of herbicides, France started at a high level, and has only recently caught up to the US. In the case of insecticides, France started out way, way higher than the US and is still way higher despite their recent reductions. Looked at this way, it's clear that France hasn't actually done that well. What's more, I had to cherry pick starting and ending years in the top set of charts in order to make France look as good as it does. Plus there's this, from Kniss:

It is true that France has been reducing pesticide use, but France still uses more pesticides per arable hectare than we do in the USA. In the case of fungicide & insecticides, a LOT more. But a relatively tiny proportion of these differences are likely due to GMOs; pesticide use depends on climate, pest species, crop species, economics, availability, tillage practices, crop rotations, and countless other factors....Given all of these confounding factors, I wonder why France was singled out by Mr. Hakim as the only comparison to compare pesticide use trends. Pesticide use across Europe varies quite a bit, and trends in most EU countries are increasing, France is the exception in this respect, not the rule. In the early 1990’s, France was using more herbicides compared to almost every other country, so it shouldn’t be too surprising that pesticide use decreased as formation of the EU began to standardize pesticide regulations after 1993.

So what's the best comparison to make: percentage change or raw volume? In this case, presenting only the percentage change leaves you with a very misleading impression. Presenting raw volumes, however, makes everything pretty clear. You can see that France started out far higher than the US and that its use of herbicides and insecticides has been decreasing. So why present only percentages?

POSTSCRIPT: If you click on the chart pack in the Times story, you will actually find charts showing raw volume of pesticide use in the US and France. However, they're shown in two different charts, using different units, and broken up into different categories. If you were deliberately trying to make a comparison nearly impossible, this is how you'd do it.

Andrew Tyndall notes that the nightly news no longer seems to care about policy debates:

This year's absence of issues is an accurate portrayal of the turf on which the election is being played out....If the candidates are not talking about the issues, the news media would be misrepresenting the contest to do so.

With just two weeks to go, issues coverage this year has been virtually non-existent.... No trade, no healthcare, no climate change, no drugs, no poverty, no guns, no infrastructure, no deficits. To the extent that these issues have been mentioned, it has been on the candidates' terms, not on the networks' initiative.

I disagree with this on two levels. First, Hillary Clinton has talked plenty about issues in the conventional sense that Tyndall means it: speeches that cover specific policy proposals, with detail to back them up. Only Donald Trump has declined to do this.

More broadly, both candidates have talked about issues. Trump talks all the time about trade, immigration, ISIS, and guns. Clinton talks about childcare, ISIS, health care, guns, and so forth. There are lots of character attacks too, but then, that's usually the case. But just because issues are talked about in broad strokes doesn't mean they're not talked about. They are. The network news broadcasts just don't want to risk losing their audiences by forcing them to pay attention to such boring stuff.

The chart below comes from Wonkblog. It's from a study of crime in Oregon, and shows that at age 21—the legal drinking age in Oregon—crime spikes considerably:

One striking chart shows how alcohol can turn people into criminals

As soon as people turned 21, their likelihood of criminality spiked considerably....The number of charges filed against 21-year-olds was similar to the number for 19-year-olds. In other words, from a criminal-justice standpoint, turning 21 is akin to turning back the clock to your late teens.

The mechanism by which this works is fairly obvious — access to alcohol increases dramatically at age 21. That brings more intoxication, and with it more aggressive, belligerent and criminally stupid behavior.

Sometimes, though, one striking chart isn't enough. Sometimes you really need to see a whole bunch of them. I apologize for the size and readability of this, but I think it's best if I show you everything, instead of just picking and choosing. Here's the complete set of charts from the Oregon study:

Virtually the entire effect is driven not by "more aggressive, belligerent and criminally stupid behavior" in general—violent crime shows no effect at all—but specifically by alcohol-related offenses: DUI/reckless driving, providing alcohol to minors, public disorder, and so forth. The authors also suggest there might be some small effect on assault, trespass, marijuana, and cocaine. But if you take a look at those charts without pre-assuming a change at age 21, you see a very vague scatterplot that doesn't really suggest anything special at that age.

Bottom line: Legal access to alcohol certainly increases alcohol use, and therefore increases the rate of drunk driving, alcohol-induced public disorder, and providing alcohol to minors. You hardly need a study to tell you that. But on all other kinds of crime? It seems to have barely any effect at all.

With a mere 6 days left in Campaign 2016, Ezra Klein points out that Hillary Clinton is perhaps the most transparent presidential candidate in history:

We have Hillary Clinton’s full tax returns going back to the year 1977...public schedules...her campaign’s donors and her foundation’s donors...tens of thousands of emails from her time at the State Department...thousands of her campaign chair’s emails...investigative reports, congressional testimony, and documentary evidence from the inquiries into Whitewater, Benghazi, and Travelgate....so many independent biographies that I couldn’t come up with an accurate count.

....The story with Trump is quite different. We have the three pages from his 1995 tax return...books Trump has written about himself...financial disclosures to the Federal Election Commission, in which he claims, in all capital letters, to have “10 BILLION DOLLARS,” but no one believes that document...Digging beyond that image is difficult because Trump has forced his former associates, and even his former romantic partners, to sign nondisclosure agreements.

Despite all this, Clinton has a reputation for opacity while Trump has a reputation for being open about everything. The reason is deceptively simple: it's what both candidates want. Clinton very clearly does her best to reveal as little as possible. Trump, by contrast, will talk about anything, loudly and volubly. It's true that when he talks, he lies constantly and says next to nothing when he's not lying, but the impression he gives is of somebody with nothing to hide.

Clinton's reputation is not unfair. Most of her openness has been forced on her, after all. Trump's reputation, by contrast, is ridiculous. He hides everything and lies about what he can't. And since he runs a private company and has never served in government, he can get away with it. He's not subject to FOIA requests or WikiLeaks dumps or random judges deciding that all his emails should be made public.

This isn't going to change, and at this point it no longer matters whether it's fair. It just is. But it's what produces such bizarre levels of CDS1 among conservatives. They've forced so much openness on Clinton in an effort to destroy her, and it drives them crazy that it's done nothing except paint a portrait of a pretty normal politician. Over 25 years, they've managed to uncover only three "scandals" that are even marginally troubling,2 and every dry well does nothing but convince them that Clinton is even more devious than they thought. By this time, we've tracked practically every hour of every day of Clinton's life for the past decade, and there's almost literally no unexamined time left. But it doesn't matter. The next one will get her for sure!

The truth is different, of course. Hillary Clinton dislikes the press and has learned to be very careful in her public utterances. She has done a few dumb things in her life, and pushed the envelope further than she should a couple of times. If you dislike her, that's fine. But basically she's a fairly ordinary politico—ironically, an unusually honest one. When she makes a deal, her word is good. When she talks about policy, she's careful not to overpromise. On the honesty front, she is Mother Teresa compared to Donald Trump.

1Clinton Derangement Syndrome, in case you've forgotten.

2The cattle futures thing remains intriguingly dodgy. Travelgate didn't involve anything illegal, but definitely shows Clinton in a bad light. And Emailgate may not have produced any evidence of wrongdoing, but it did uncover a case of poor judgment.

The Wall Street Journal always has the cheeriest news:

Banks no longer reign over the mortgage market. They accounted for less than half of the mortgage dollars extended to borrowers during the third quarter....Taking their place are nonbank lenders more willing to make riskier loans banks now shun.

....Many nonbanks are courting borrowers who can’t get approved by banks, which have favored customers with pristine credit....As nonbanks get bigger, some analysts are concerned about their ability to weather tougher economic times. Unlike banks, nonbanks don’t take in deposits and rely largely on financing—mostly from banks—to keep loan volume going.

....Some nonbank lenders say concerns about liquidity are overblown. “We are not taking on credit risk—the loss is borne by [government] agencies,” said Bob Walters, chief economist at Quicken Loans. “As long as there is a government guarantee, that is a powerful leveling factor that keeps the flow of funds going.”

On the one hand, it's not like I want to panic every time home mortgages become marginally easier to get. On the other hand, well, you know. The Late Unpleasantness™ continues to weigh on my mind. This is why I take seriously Hillary Clinton's promise to regulate shadow banking more rigorously. If we pass laws that tighten up the banking industry, the point is to tighten up the industry, not merely to shift risky business somewhere else.

So, sure, there's probably no need to melt down over every minor change in the mortgage market. But it's sure as hell worth paying attention to.

I mentioned yesterday that I remain pretty relaxed about the latest presidential polls. To explain this, I'd like to nominate two phrases of the year: response rates and reversion to the mean.

Response rates

I haven't written about this before, but it's been a hot topic of conversation in polling circles this year. In a nutshell, lots of pollsters have come to believe that voters are less likely to respond to polls when their candidate does badly. YouGov, which uses a panel-based polling system that allows them to re-interview people, says this:

After the first presidential debate in September, we reinterviewed 2,132 people who had told us their vote intentions a month before. 95 percent of the September Clinton supporters said they intended to vote for her....Of the Trump supporters, only 91 percent said they were still planning on voting for Trump....The net effect was to increase Clinton’s lead by almost four points. That was real change, though significantly less that the ten point change to Clinton’s lead seen in some polls....Other events, however, have not had any detectable impact on voting intentions. We did not see any shifts after the release of the Access Hollywood video, the second or third presidential debates, or the reopening of the FBI investigation into Clinton’s emails.

....Although we didn’t find much vote switching, we did notice a different type of change: the willingness of Clinton and Trump supporters to participate in our polls varied by a significant amount depending upon what was happening at the time of the poll: when things are going badly for a candidate, their supporters tend to stop participating in polls. For example, after the release of the Access Hollywood video, Trump supporters were four percent less likely than Clinton supporters to participate in our poll. The same phenomenon occurred this weekend for Clinton supporters after the announcement of the FBI investigation.

The chart on the right shows what happened to YouGov's response rates after the second debate: people who had voted for Obama in 2012 answered the phone more, while people who had voted for Romney answered the phone less. This produces a swing in the final poll results, but it's meaningless. The same people are still out there, and they're mostly still planning to vote for the same person.

The whole YouGov piece is worth a read, because it does a good job of explaining what this phenomenon is all about, as well as offering possible solutions. But the bottom line is simple: A lot of the poll swings we see in presidential campaigns are probably illusions, demonstrating only changes in the willingness to be polled, not changes in voting intention. If this gets confirmed, it represents a genuine sea change in how we interpret polls.

Reversion to the mean

This theory about response rates might explain another phenomenon that's been much in evidence this year: reversion to the mean. You can see it pretty clearly in Sam Wang's meta-margin:

Up through July, Hillary Clinton was ahead of Trump by a steady 3.5 points. Then she got a big spike after the Democratic convention, but it quickly reverted to around 3.5. Then she dropped a bunch after some email news and her fainting spell on 9/11—but again, the trendline quickly reverted to 3.5. Then she spiked again after the second debate and the Access Hollywood video, but reverted to 3.5 yet again.

Clinton could easily lose another point before Election Day, or she could revert back to 3.5 and stay there. I'd bet on reversion to the mean. This election features two candidates who have been around a long time and are both very well known. Almost everybody made up their minds pretty early, and nothing much has changed for the past 12 months. Hillary Clinton will most likely win by 3-4 percentage points, plus maybe a little extra because she has a way better ground game.

Someone at the Washington Post thinks they're being awfully cute today:

They just don't get it. Any congressional session held under a Trump presidency is special. Very, very special. In fact, the most special session we've ever held. What's so hard to understand about that?

The latest hotness on the right is to promise not just to hold up Senate hearings on Merrick Garland until we get a new president, but to hold up all hearings for all Supreme Court nominees forever if Hillary Clinton wins:

That prospect — which could impact every aspect of American life including climate regulations, abortion and gun rights — was first raised by Senator John McCain of Arizona, then Ted Cruz of Texas and now Richard Burr of North Carolina, who CNN reported Monday talked up the idea at a private event over the weekend.

“If Hillary Clinton becomes president, I am going to do everything I can do to make sure four years from now, we still got an opening on the Supreme Court,” Burr, chairman of the Senate Intelligence Committee, told a group of Republican volunteers, according to CNN.

Marco Rubio, taking his usual craven approach to political landmines, says it would be wrong to blockade everyone, but it would be OK to blockade anyone who's not a conservative:

“If it’s someone good who understands that their job is to apply the constitution, according to its original intent, then that will be a welcome surprise,” he said. “But barring whether it’s Republican or a Democrat, if they appoint someone who I believe doesn’t meet that standard I’ll oppose that nominee.”

Ross Douthat explains the principled thinking behind this strategy:

There you have it. Liberal views of the law are inherently illegitimate, so Democrats don't get to pick any more Supreme Court justices. There's a name for this kind of republic. Starts with a B. Not quite coming to me, though.