The NSA isn't allowed to spy on Americans, but the nature of modern communication doesn't always make it obvious whether a phone call or email is foreign or domestic. This means that in the course of its normal business of spying on foreigners, NSA will inevitably collect information it shouldn't have. Certain rules, called "minimization procedures," define what NSA is required to do when it discovers that it has inadvertently captured a U.S. person in its surveillance dragnet.

Today, in the latest release of classified NSA documents from Glenn Greenwald, we finally got a look at these minimization procedures. Here's the nickel summary:

The top secret documents published today detail the circumstances in which data collected on US persons under the foreign intelligence authority must be destroyed, extensive steps analysts must take to try to check targets are outside the US, and reveals how US call records are used to help remove US citizens and residents from data collection.

I have a feeling it must have killed Glenn to write that paragraph. But on paper, anyway, the minimization procedures really are pretty strict. If NSA discovers that it's mistakenly collected domestic content, it's required to cease the surveillance immediately and destroy the information it's already collected. However, there are exceptions. They can:

Retain and make use of "inadvertently acquired" domestic communications if they contain usable intelligence, information on criminal activity, threat of harm to people or property, are encrypted, or are believed to contain any information relevant to cybersecurity.

The Guardian has posted two classified documents online. The first one describes the procedure for determining whether a surveillance target is legitimate (i.e., a non-U.S. person located outside the country). The second one describes the minimization procedures in case of inadvertent targeting of a U.S. person. There are a few obvious things to say about them:

  • The determination document repeatedly emphasizes that NSA bases its decisions on the "totality of the circumstances." There are quite a few safeguards listed to make sure that only foreigners are targeted, but in the end these are often judgment calls from analysts.
  • The minimization procedures are fairly strict, but they do allow retention and dissemination of domestic data—without a warrant—under quite a few circumstances. "Threat of harm" is pretty broad, as is "criminal activity." The latter, in fact, seems like a loophole the size of a Mack truck. It suggests that NSA could have a significant incentive to "inadvertently" hoover up as much domestic information as possible so it can search for evidence of criminal activity to hand over to the FBI.
  • The oversight procedures are pretty thin. Analysts have quite a bit of discretion here.

It's genuinely unclear how big a problem this stuff is. It's plainly true that determining whether someone is a U.S. person is sometimes a judgment call, and it's possible that mistakes are rare. What's more, if collection of domestic content genuinely is inadvertent, and is only occasionally turned over to other agencies when there's evidence of serious crime, we should all feel better about this. But we really have no way of knowing. That would require, say, an inspector general to gather this kind of information, and the IG has specifically declined to do this.

Also, note that the documents posted by the Guardian are from 2009. It's quite possible that procedures have changed since then.

This is really a helluva mystery. A new Gallup poll shows that 87 percent of Americans favor a path to citizenship for undocumented immigrants who are currently in the country. Only 12 percent oppose it. So where's the opposition to comprehensive immigration reform coming from? I realize that intensity of feeling matters, but it doesn't matter much when the opposition is so minuscule. If Gallup's numbers are right, even a majority of tea partiers favor a path to citizenship.

So what's going on here? Why are House Republicans apparently feeling so much pressure over this?

As near as I can remember, I've been reading about our unsustainable shortage of primary care physicians since—oh, since about when I learned to read. And I always sort of vaguely wondered why the law of supply and demand seemed to have broken down here. If GPs are in short supply and specialists are a glut on the market, then salaries should go up for GPs and down for specialists. Eventually the market will get to the right equilibrium. Right?

But that hasn't happened. And not for lack of time. According to Phil Longman, in a new piece for the Washington Monthly, the problem of overspecialization has been recognized since the early 1960s, which means that my joke about it being around since I learned to read is just about accurate. (Except for the part about the five-year-old Kevin Drum actually reading about it.) So what's up? Why are specialists still paid so much if there are too many of them?

Longman says there are two problems. First, Medicare sets reimbursement rates, and thanks to the stranglehold that specialists have on the process via an advisory panel called RUC, those rates have stayed high regardless of market forces. But that's only been the case since 1991, and the problem goes back further than that. So again, what's up?

That brings us to Longman's second problem: America's residency programs reserve lots of slots for specialists and not very many for GPs. So regardless of market forces, we get more specialists than we need or want and not enough GPs. But the underlying problem remains Medicare, which subsidizes residency programs, and the stranglehold that specialists have on its residency funding program:

As an indication of just how perennial this behind-the-scenes struggle has been, in 1980 a federal advisory panel on graduate medical education was already criticizing the system for pumping out an excess supply of specialists. By 1985, none other than then Senator Dan Quayle was penning an article for the policy journal Health Affairs calling for Medicare to stop subsidizing residency programs that didn’t send at least 70 percent of their graduates into primary care. By 1989, the Institute of Medicine was weighing in, saying that graduate medical education programs were too concentrated in hospitals and were failing to provide proper training for primary care physicians. Yet despite the alliance of conservatives alarmed by the mounting cost of Medicare and reformers intent on improving the practice of American medicine, nothing changed, thanks to the entrenched powers of specialists within both academic medicine and the health care system generally.

I'll confess that I'm still a little baffled. If I were, say, a radiologist, the last thing I'd want is an army of young radiology specialists flooding the market every year. Sure, it's nice that my fellow radiologists on the RUC will try to keep my payments high no matter what, but wouldn't it be better to rely on supply and demand? That seems more reliable than a finger in the dike that could get pulled out if Congress ever gets a bee in its bonnet. And restricting supply is hardly a novel idea. It's pretty much how every other occupational licensing body in the world works.

So I continue to be a little puzzled about this. Still, the whole piece is worth a read for a look into how regulatory capture works and what it might take to fix things. And for your further amusement, the table on the right is Longman's Hall of Shame: the 20 worst teaching programs in the country in terms of producing primary care physicians. Go boo them the next chance you get.

Yesterday Ben Bernanke kinda sorta said that the Fed might taper off its bond buying program slightly earlier than expected. Maybe. Naturally, markets went nuts. I ignored it, because I figured it was just Wall Street doing what Wall Street does: panicking over everything. Either that or the robots that do most of the trading were engaged in a few trillion epic microsecond battles. But it would all settle down before long.

As it turns out, investors might have been reacting to bad news out of China as much as they were to Bernanke. Who knows? In any case, stock markets fell, interest rates rose, and commodity prices dropped. No surprise there. But Neil Irwin points out something else: inflation expectations plummeted. Not right away, though. Not until this morning, when five-year inflation expectations dropped sharply from 1.86 percent to 1.76 percent.

So what caused this? If it were Bernanke, it would have happened yesterday. Is it generalized fear that the slowdown in China is going to hurt the global economy, which will lead to lower inflation in the U.S.? A reaction to the rise of the dollar? It's puzzling—and the size of the drop is puzzlingly large and sharp. What's worse, the Keynesians and the market monetarists are sure to provide us with diametrically opposite explanations of why this happened, which will keep us all even more puzzled.

But that said, current inflation is under control and inflationary expectations—for whatever reason—are obviously pretty firmly under control too. This is exactly the opposite of what we want, so it's bad news no matter what the reason.

The New York Times has an intriguing interview today with Laszlo Block, Google's head of HR senior vice president of people operations. He says that most job interviews are basically a waste of time: Google has gathered mountains of data and discovered that there's "zero relationship" between how job candidates are scored by hiring managers and how well they eventually do on the job. Also: GPA and test scores are pretty much useless if you've been out of school more than a few years. That shouldn't come as a surprise. I've never understood why anyone would care even slightly about that stuff once you've got some real-world job experience to evaluate.

But what about those famous Google brainteasers? Those are great, right? Not so much:

On the hiring side, we found that brainteasers are a complete waste of time. How many golf balls can you fit into an airplane? How many gas stations in Manhattan? A complete waste of time. They don’t predict anything. They serve primarily to make the interviewer feel smart.

Instead, what works well are structured behavioral interviews, where you have a consistent rubric for how you assess people, rather than having each interviewer just make stuff up.

Behavioral interviewing also works — where you’re not giving someone a hypothetical, but you’re starting with a question like, “Give me an example of a time when you solved an analytically difficult problem.” The interesting thing about the behavioral interview is that when you ask somebody to speak to their own experience, and you drill into that, you get two kinds of information. One is you get to see how they actually interacted in a real-world situation, and the valuable “meta” information you get about the candidate is a sense of what they consider to be difficult.

Actually, this advice has been conventional wisdom for quite a while among people who know what they're talking about. But it's hard! And no one likes to do it. Most people are convinced that they have a mystical ability to evaluate others just by chatting with them and "sizing them up." Well, guess what? You probably don't. And you especially don't in the very formalized setting of a job interview.

Still, this belief is based on hundreds of centuries of evolved human nature. I don't expect it to change anytime soon.

The Benghazi "scandal" has remained a conservative pet rock for far longer than anyone had any right to expect. One reason is that it's become a game of whack-a-mole: you debunk one "stand down" conspiracy theory and another takes its place. First it was a fast-response team that President Obama refused to deploy. Then it was an enterprising colonel who was relieved of duty for getting to ready to send in a team even though he had orders not to. Then it was a team in Tripoli. Then a C-110 team in Croatia. Then it was yet another team in Tripoli. Basically, the facts didn't matter. The tea party crowd had convinced itself that Obama was a craven political schemer who refused to mount a rescue of American troops in danger, so they figured that one of these stories just had to be true.

Today, Dave Weigel says that more or less the same thing is going on with the IRS scandal. Conservatives just know that somehow this all came from the top. It happened because Obama was in a panic over the power of the tea party movement, so he put out orders to crush it. In some tellings—this is the "culture of intimidation" story—he merely made his wishes known and the bureaucracy leaped to obey. In other versions he personally met with the IRS director 157 times to make sure his plans were being carried out. Other stories implicate his top staff. An even newer version veers away from harassment of tea party groups over their tax exempt status—which, let's face it, is about the least effective way imaginable of trying to silence them—and is now focused on Obama's even more insidious campaign to directly audit and torment important Republican donors.

Today, Stan Veuger of AEI, a recently minted PhD apparently trying to make a name for himself, presents yet another theory. Tea party groups, he says, presented an "existential threat" to Obama's reelection, and it turns out that failing to approve their tax-exempt status "would have been an effective campaign strategy going into the 2012 election cycle." His paper examines how rainfall affected attendance at tea party rallies, and how that in turn affected Republican turnout in the 2010 election. If you then assume that 2012 was just like 2010, and that jitters over tax status dramatically affected tea party effectiveness, it turns out that Obama stole the 2012 election from Romney! Specifically, "had their effect on the 2012 vote been similar to that seen in 2010, they would have brought the Republican Party as many as 5 - 8.5 million votes compared to Obama's victory margin of 5 million." Weigel calls this out for the nonsense it is:

OK, starting with the general agreement that the IRS shouldn't have leaned so hard on the Tea Party, let's further agree that this is madness. Leave aside the fact that most real political activity credited to the "Tea Party" was organized by groups—FreedomWorks, Americans for Prosperity—that have had favorable tax status for years. From January 2009 to November 2010, the Tea Party went from zero to more than thousand groups. From December 2010 to November 2012, the movement was growing less popular for reasons that had nothing to do with individual groups' tax status. A November 2010 Quinnipiac poll gave the Tea Party a mildly negative 34-38 favorable rating; a year later this had fallen to a 29-42 favorable rating. Politics and life had intervened, and the "Tea Party's" role in the debt limit showdown (i.e., starting the showdown in the first place) was wildly unpopular. Simultaneously, a slowly growing economy was making the president more popular.

So why assume that the movement would have continued to grow at its 2009-2010 pace even after the environment for its growth had been radically altered? 

Why indeed? It doesn't even remotely make sense. But it's good enough for AEI! I fully expect this study to be the lead segment on Hannity tonight. As Weigel says, "We're not even close to the end of IRS scandal-oriented theories of liberal treachery."

The interbank lending rate in China has spiked over the past couple of weeks. "China's interbank market is basically frozen — much like credit markets froze in the United States right after Lehman failed," Patrick Chovanec told the New York Times. Is this because China's economy has overheated and its credit bubble is now bursting? Maybe. But the Financial Times says there's more to it:

In the midst of the extreme market stress, a statement issued late Wednesday by the central bank raised the possibility that politics are also playing an important role.

Bankers had been calling for the central bank to ease the pressure and a few investors had even predicted that it might cut interest rates. Instead, the People’s Bank of China ordered a thorough implementation of the new “mass line education” campaign launched this week by President Xi Jinping — a campaign that in its propaganda-style and potential scope carries echoes of the Mao era.

The Communist party cadres that run the central bank were told to attack the “four winds” of “formalism, bureaucracy, hedonism and extravagance”, as demanded by Mr Xi.

....In monetary policy terms, the central bank could certainly be said to be waging war on hedonism and extravagance. The seven-day bond repurchase rate, a key gauge of liquidity in China, surged 270 basis points to more than 10.8 per cent on Thursday — a punitively high rate that could force cash-hungry banks to call in the riskiest of their loans.

Andrew Mellon would be proud! But if this really is what Xi is doing, he'd better be careful. As you may recall, things didn't turn out the way Mellon hoped.

Tyler Cowen points out that Greece is once again in crisis, this time over a dispute about shutting down its public broadcasting outlet:

The current Turkish crisis was set off by a dispute over a public park, and the recent demonstrations in Brazil seem to have been prompted by a 7% hike in bus fare prices, which is about ten U.S. cents....Many deconversions from religion, or from fandom, or even from marriage, work the same way. Big lies are told and those lies inflict some damage. The institution in question soldiers on. A bit later, an apparently smaller slight or problem brings the whole thing crashing to the ground, precisely when things appeared to be getting better.

I’m not saying it always runs that way, only that it is a very common path. Furthermore the steepest period of decline is very often when people are too preoccupied with coping to make the major adjustment.

The bottom line is that one should not dismiss the importance of small events, especially these days.

This is one of the things that makes foreign policy forecasts so inherently unreliable. Sometimes a small event is just a small event, and it peters out after a few days or weeks. Other times it's like a snowball rolling downhill. Franz Ferdinand wasn't that important to the future of the Austro-Hungarian empire, after all.

In any case, this is how I've always viewed the euro crisis ending. Things will roll along, with patch after patch calming things down, until someday a metaphorical archduke is assassinated. At that point, either everyone finally gets scared enough to really do something about the fundamentals, or else the eurozone spins apart. But the when, how, and whether are pretty much impossible to predict.

Several people have suggested that President Obama will make climate change a key initiative of his second term. I've never really believed that, but today the New York Times reports that it might be for real:

President Obama is preparing a major policy push on climate change, including, for the first time, limits on greenhouse gas emissions from new and existing power plants, as well as expanded renewable energy development on public lands and an accelerated effort on energy efficiency in buildings and equipment, senior officials said Wednesday

Heather Zichal, the White House coordinator for energy and climate change [...] suggested in her remarks that a central part of the administration’s approach to dealing with climate change would be to use the authority given to the Environmental Protection Agency to address climate-altering pollutants from power plants under the Clean Air Act. She said none of the initiatives being considered by the administration required legislative action or new financing from Congress.

The EPA can actually do a fair amount if it decides to. And Republicans know it: it's one of the reasons they've held up the nomination of Gina McCarthy to head up the EPA. This announcement is likely to turn up the heat in that battle another notch or two.

Chutzpah awards are really getting hard to hand out these days. I just gave Darrell Issa one, but now I see that Sen. Jeff Sessions provided this explanation yesterday of why he opposes immigration reform even though the CBO says it would be good for the economy:

This increased GDP will be at the expense of poor and working-class Americans. The benefit will go to the business owners while the wages of U.S. workers—which should be growing—will instead decline

Um....since when has Jeff Sessions had a problem with benefits flowing to business owners? And since when has he demonstrated even the slightest concern with the fortunes of the poor? Since never. But I guess people can evolve on these things, so maybe we're now seeing a new, more compassionate Jeff Sessions. Maybe.

Ezra Klein has more details if you can stomach them.