Kevin Drum

SIM Card Manufacturer Says Its Encryption Keys Are Safe From NSA Hacking

| Wed Feb. 25, 2015 12:19 PM EST

I'm passing this along without comment since I don't have anything substantive to add. I just wanted to keep everyone up to date on the Intercept story about the NSA stealing cell phone encryption data stored on SIM chips:

Security-chip maker Gemalto NV said Wednesday that American and British intelligence services could be responsible for a “particularly sophisticated intrusion” of its networks several years ago, but denied that the alleged hack could have widely compromised encryption it builds into chips used in billions of cellphones world-wide.

....Company executives also asserted that the interceptions wouldn’t have compromised the security of its newer SIM cards for 3G and 4G cellular networks, only older 2G networks. The reason: Gemalto says the new technology no longer require it to send telecom companies the keys to decrypt individuals’ communications—so they couldn’t have been intercepted.

Hmmm. On the one hand, many of the Snowden documents are indeed fairly old, dating back to 2010 or 2011. So they could be out of date. On the other hand, the NSA didn't necessarily have to "intercept" anything here. A sufficiently sophisticated hack could presumably have given them direct access to the Gemalto database that contains the encryption keys. And needless to say, Gemalto has a vested interest in assuring everyone that their current products are safe.

So....who knows what really happened here. We'll likely hear more about it as Gemalto's internal investigation continues.

Advertise on

DHS Funding Fight Is Going Down to the Wire

| Wed Feb. 25, 2015 11:31 AM EST

We're getting down to the wire in the funding fight over the Department of Homeland Security: DHS will shut down this weekend if funding isn't approved by Friday. In the Senate, Mitch McConnell wants to simply hold two separate votes: one to fund DHS and another to repeal President Obama's recent immigration actions. But tea partiers in the House are adamantly opposed to that: they want to keep the two things together in one bill, which they hope will force Democrats to cave in and kill the immigration plan. In reality, it will only produce deadlock in the Senate and a shutdown of DHS that Republicans will be blamed for. So what's John Boehner to do? Greg Sargent comments:

We’ve seen this particular thriller a number of times already. Here’s how it always goes: We are told there’s no way Boehner would ever dare move must-pass legislation with a lot of Democrats. He’s stuck! Then pressure builds and builds, and Boehner does end up passing something with a lot of Democrats. Last I checked, he’s still Speaker.

....The fact that Boehner has the mere option of passing clean funding with the help of a lot of Democrats is rarely even mentioned. You can read article after article about this whole showdown and not be informed of that basic fact. Thus, the actual reason we’re stuck in this crisis — Boehner is delaying the moment where he does pass something with Dems for as long as possible — goes oddly unmentioned. Yet recent history suggests that Boehner himself knows this is how it will end, and that all of this drama won’t change the outcome.

Probably so. After all, the only thing that changed in the last election was control of the Senate, and Senate Republicans are willing to compromise. The House is probably going to have to go down that road eventually too.

But my guess is that they're going to shut down DHS for a while first. Boehner has made it pretty clear that he feels like he needs to demonstrate his conservative bona fides at the beginning of this new session of Congress, and that means holding out as long as he can. It's a waste of time, and it's going to hurt Republican efforts to work on other legislation, but that's life. Symbols are important, and Boehner needs to show whose side he's on. There's a good chance this will last a couple of weeks before it gets resolved.

Everything You've Been Told About Healthy Eating Is Wrong, Except This

| Tue Feb. 24, 2015 3:59 PM EST

For several years now I've been following the controversy over whether the dietary guidelines that have developed over the the past 70 years might be all wrong. And I've become tentatively convinced that, in fact, they are wrong. For most people—not all!—salt isn't a big killer; cholesterol isn't harmful; and red meat and saturated fat are perfectly OK. Healthy, even. Sugar, on the other hand, really needs to be watched.

Before I go on, a great big caveat: I'm not even an educated amateur on this subject. I've read a fair amount about it, but I've never dived into it systematically. And the plain truth is that firm proof is hard to come by when it comes to diet. It's really, really hard to conduct the kinds of experiments that would give us concrete proof that one diet is better than another, and the studies that have been done almost all have defects of some kind.

In other words, what follows are some thoughts I've gathered over the years, not a crusade to convince you I'm right. And it's strictly about what's healthy to eat, not what's good for the planet. Take it for what it's worth.

Salt is perhaps the most personal subject to me. My father had a stroke when I was a teenager, and his doctor told him he needed to watch his salt intake. Ever since then, I've watched mine too. As it happens, this wasn't a big sacrifice: I don't eat a lot of prepared foods, which are usually loaded with salt, and I've never felt the need to heavily salt my food.

Nevertheless, last year my doctor told me she was worried about my sodium level. I misunderstood at first, and figured that I needed to make additional efforts to cut back. But no. My serum sodium level was too low. What's more, it turns out that most Americans consume a safe amount of sodium. The usual recommendation is to keep sodium intake below 2400 mg per day, but the bulk of the evidence suggests that twice this much is perfectly safe for people who don't suffer from hypertension. (And even the recommendations for people with hypertension might be more restrictive than they need to be.)

Then there's cholesterol. I guess I don't have to say much about that: the evidence is now so overwhelming that even the U.S. government's top nutrition panel announced a couple of weeks ago that dietary cholesterol was no longer a "nutrient of concern" in its latest guidelines. Go ahead and have an egg or three.

Finally, there's saturated fat. The same nutrition panel that decided cholesterol is OK didn't ease up its recommendations on saturated fat. But I'm increasingly skeptical of this too. Interestingly, Aaron Carroll is skeptical too:

As the guidelines have recommended cutting down on meat, especially red meat, this meant that many people began to increase their consumption of carbohydrates.

Decades later, it’s not hard to find evidence that this might have been a bad move. Many now believe that excessive carbohydrate consumption may be contributing to the obesity and diabetes epidemics. A Cochrane Review of all randomized controlled trials of reduced or modified dietary fat interventions found that replacing fat with carbohydrates does not protect even against cardiovascular problems, let alone death.

Interestingly, the new dietary recommendations may acknowledge this as well, dropping the recommendation to limit overall fat consumption in favor of a more refined recommendation to limit only saturated fat. Even that recommendation is hotly contested by some, though.

....It is frustrating enough when we over-read the results of epidemiologic studies and make the mistake of believing that correlation is the same as causation. It’s maddening, however, when we ignore the results of randomized controlled trials, which can prove causation, to continue down the wrong path. In reviewing the literature, it’s hard to come away with a sense that anyone knows for sure what diet should be recommended to all Americans.

Randomized trials are the gold standard of dietary studies, but as I said above, they're really, really hard to conduct properly. You have to find a stable population of people. You have to pick half of them randomly and get them to change their diets. You have to trust them to actually do it. You have to follow them for years, not months. Virtually no trial can ever truly meet this standard.

Nonetheless, as Carroll says, the randomized trials we do have suggest that red meat and saturated fat have little effect on cardiovascular health—and might actually have a positive effect on cancer outcomes.

At the same time, increased consumption of sugars and carbohydrates might be actively bad for us. At the very least they contribute to obesity and diabetes, and there's some evidence that they aren't so great for your heart either.

So where does this leave us? As Carroll says, the literature as a whole suggests that we simply don't know. We've been convinced of a lot of things for a long time, and it's turned out that a lot of what we believed was never really backed by solid evidence in the first place. So now the dietary ship is turning. Slowly, but it's turning.

For myself, I guess I continue to believe that the key is moderation. Try to eat more fresh food and fewer packaged meals. That said, there's nothing wrong with salt or saturated fat or cholesterol or sugar. None of them need to be cut down to minuscule levels. You don't need to limit yourself to two grams of salt or eliminate red meat from your diet. You can eat eggs and butter and steak if you want to. You should watch your sugar and carb intake, but only because so many of us consume truly huge quantities of both. In the end, all of these things are OK. They simply need to be consumed in moderation.1

Can I prove that? Nope. But it's what I believe these days.

1Needless to say, none of this applies to people with specific conditions that require dietary restrictions. Listen to your doctor!

Red Barns and White Barns: Why Rural Crime Skyrocketed in the Late 1800s

| Tue Feb. 24, 2015 1:51 PM EST

Here's a fascinating little anecdote about lead and crime from a recent paper by Rick Nevin. It shouldn't be taken as proof of anything, but it's certainly an intriguing little historical tidbit about the association between lead exposure and increases in crime rates.

Here's the background. Homicides increased dramatically between 1900-11, but most of that appears to be the result of increased rural homicides, not urban homicides. If lead exposure is part of the reason, it would mean that rural areas were exposed to increasing levels of lead about 20 years earlier, around 1880 or so. But why? Nevin suggests that the answer to this question starts with another question: Why are barns red?

Professional painters in the 1800s prepared house paint by mixing linseed oil with white lead paste. About 90% of Americans lived in rural areas in the mid-1800s, and subsistence farmers could make linseed (flaxseed) oil, but few had access to white lead, so they mixed linseed oil with red rust to kill fungi that trapped moisture and increased wood decay. Red barns are still a tradition in most USA farming regions but white barns are the norm along the path of the old National Road. Why?

....The reason the red barn tradition never took root along that path is likely because the National Road made freight, including white lead, accessible to nearby farmers. USA lead output was a relatively stable 1000 to 2000 tons per year from 1801-1825, but lead output was 15,000 to 30,000 tons per year from the mid-1830s through the mid-1860s after the completion of the National Road.

....The first American patent for “ready-mixed” paint was filed in 1867; railroads built almost 120,000 track miles from 1850 to 1900; and Sears Roebuck and other mail-order catalogs combined volume buying, railroad transport, and rural free parcel post delivery to provide economical rural access to a wide variety of products in the 1890s.

The murder arrest rate in large cities was more than seven times the national homicide rate from 1900-1904 because lead paint in the 1870s was available in large cities but unavailable in most rural areas. The early-1900s convergence in rural and urban murder rates was presaged by a late-1800s convergence in rural and urban lead paint exposure.

In short, lead paint simply wasn't available in most rural areas before the 1880s except in very narrow corridors with good transportation. You can see this in the prevalence of white barns along the National Road. Then, starting in the 1880s, revolutions in both rail transport and mail order distribution made economical lead paint available almost everywhere—including rural areas. A couple of decades later, homicide rates had skyrocketed in rural areas and had nearly caught up to urban murder rates.

By itself, of course, this would be merely speculative. What makes it more than this is that it adds to the wealth of other evidence that lead exposure in childhood leads to increased violence in adulthood. In the post-World War II era, lead exposure came mainly from automobile exhausts, but in the post-Civil War era it came mainly from the growth in the use of lead paint. And when lead paint became available in rural areas, farmers found it just as useful as everyone else. Given what we now know about the effects of lead, it should come as no surprise that a couple of decades later the murder rate in rural areas went up substantially.

There's much more in the full paper, including another question: why did murder rates in St. Louis increase 10-fold from 1910 to 1916? Can you guess the answer? I'll bet you can.

What the Broadband Industry Really Needs Isn't Net Neutrality. It Needs Competition.

| Tue Feb. 24, 2015 11:20 AM EST

Will strong net neutrality rules reduce the incentive for cable companies to invest in high-speed network infrastructure? Maybe, though similar rules certainly haven't had that effect in the cell phone market. Of course, the cell phone market is intensely competitive, and that's probably the real difference between the two. As Tim Lee notes today, Comcast's cable division is immensely profitable—certainly profitable enough to fund plenty of new high-speed infrastructure. But why should they bother?

Comcast's high profits are evidence of high barriers to entry in the broadband industry. Ordinarily, a company that consistently made billions of dollars in profits would attract new competitors seeking to capture a piece of the market.

But with a few exceptions — such as Google's projects in Kansas City and elsewhere — this hasn't really happened. In most parts of Comcast's service territory, consumers' only alternative for broadband service is the local phone company.

Conversely, Comcast doesn't seem interested in trying to steal market share from rivals. Comcast could expand into the service territory of neighboring cable companies or it could spend money building a next-generation fiber optic network the way Verizon and Google have done. Instead, they've chosen to spend more money rewarding shareholders than investing in their networks.

Given current political realities, strong net neutrality rules are a good idea. But an even better idea would be to forget about net neutrality and open up local markets to real competition. I think we'd find out pretty quickly that broadband suppliers have plenty of money for infrastructure upgrades if the alternative is a steadily shrinking market share as competitors start eating their lunch.

Competition is good. Big companies don't like it, and our approach to antitrust enforcement has unfortunately lost sight of competition as a sufficient raison d'être. That's too bad. It's the cure for a lot of ills and a way to keep the rest of the regulatory state relatively light. It's well past time for us to rediscover this.

Chart of the Day: Here's Who's Defaulting on Student Debt

| Tue Feb. 24, 2015 11:01 AM EST

Alex Tabarrok passes along the chart on the right, which shows the default rate on student loans. What it shows is surprising at first glance: the highest default rates are among students with the lowest debt, not the highest.

But on second glance, this isn't surprising at all. I'd suggest several good reasons to expect exactly this result:

  • The very lowest debt levels are associated with students who drop out after only a year or so. They have the worst of all worlds: only a high school diploma and a low-paying job, but student debt that's fairly crushing for someone earning a low income.
  • The next tier of debt is likely associated with students at for-profit trade schools. These schools are notorious for high dropout rates and weak job prospects even for graduates.
  • The middle tier of debt levels is probably associated with graduates of community colleges and state universities. Graduates of these schools, in general, get lower-paying jobs than graduates of Harvard or Cal.
  • Conversely, high debt levels are associated with elite universities. Harvard and Cal probably have pretty high proportions of students who earn good incomes after graduation.
  • The highest debt levels are associated with advanced degrees. The $50,000+ debt levels probably belong mostly to doctors, lawyers, PhDs, and so forth, who command the highest pay upon graduation.

A commenter suggests yet another reason for high default levels at low levels of debt: it's an artifact of "students" who are already deep in debt and are just looking for a way out: "The word is out if you have bad credit and are desperate for funds just go to a community college where tuition is low and borrow the maximum....Want the defaults to go down — stop lending to students that have a significant number of remedial courses their 1st and 2nd terms at a college where tuition is already low."

If you're likely to complete college, student loans are a good investment. But if you're right on the cusp, you should think twice. There's a good chance you'll just end up dropping out and you'll end up with a pile of student loans to pay back. If you're in that position, think hard about attending a community college and keeping student loans to the minimum you can manage.

And try majoring in some field related to health care. Occupations in health care appear to have a pretty bright future.

Advertise on

Quote of the Day: "I Am Coming After You With Everything I Have"

| Tue Feb. 24, 2015 12:56 AM EST

From Bill O'Reilly, to a reporter who called to ask about a Mother Jones report that he had wildly exaggerated his coverage of the Falklands War:

During a phone conversation, he told a reporter for The New York Times that there would be repercussions if he felt any of the reporter’s coverage was inappropriate. “I am coming after you with everything I have,” Mr. O’Reilly said. “You can take it as a threat.”

Charming, as always. And once again, this is the difference between O'Reilly and Brian Williams. O'Reilly and Fox News will never admit any wrongdoing, and will fight back with everything they've got. There will be no six-month suspension for Bill O'Reilly.

Will it work? Probably yes. After all, O'Reilly is paid to be a windbag, so the fact that he's exaggerated some stuff on his personal resume seems like it's just part of the package. Still, I admit that this episode is getting a lot more attention than it was when I first commented on it. The fact that the New York Times is covering it on its front page is proof of that. So maybe it's going to hurt O'Reilly more than I thought. Stay tuned.

Once Again: What's the Deal With the Pretense That the Academy Awards Are Supposed to Last 3 Hours?

| Mon Feb. 23, 2015 4:41 PM EST

Kelsey McKinney writes today about why Joan Rivers was left out of the "In Memoriam" segment at the Oscars last night:

The sequence, ultimately, only has so much room. Every year dozens of Academy Award nominees die, but there's only room to memorialize about 30 of them in a show that almost always runs over time already.

Whoa. Hold on. The Academy Awards almost never run over time. They are, quite plainly, expected to last 3½ hours. For one thing, they always last 3½ hours.1 For another, there's abundant evidence that show directors know exactly how long each bit is going to last. And there's also the evidence of other awards shows, which demonstrates that directors can hit a scheduled end mark within a minute or two. Every time. So they know perfectly well that the Oscar telecast is going to last 3½ hours.

But for some reason, the publicly acknowledged length of the show is 3 hours. Why? I've asked this before. It can't be too deep a secret since it's so obviously planned this way and has been for years. But why?

1Actually this year they really did run long, a little over 3 hours and 35 minutes. But that's unusual.

New Retirement Regs Might Pose a Campaign Problem for Republicans

| Mon Feb. 23, 2015 2:01 PM EST

Congress is now controlled by Republicans, and it's unlikely they're going to pass any of the items on President Obama's agenda. But what about executive actions? Are there any more of those left in Obama's toolkit?

Jared Bernstein says yes. Forty years ago, when rules were set regarding retirement programs, most retirement funds were managed by corporations or unions, and it was assumed that the fund managers were financially sophisticated. This meant the rules could be fairly light. But that's obviously changed: most pensions these days are IRA and 401(k) accounts that are managed by individuals who often have a hard time telling good advice from bad:

The result was a lot of people without a lot of investment acumen trying to wade through thickets of annuities, bonds, securities, and index funds, often guided by advisors and brokers who they assumed were wholly on their side.

Many were — but research shows that many were, and are — not always acting in their clients’ best interest, generating unnecessary fees and charges that erode retirement savings. The newly proposed rule, which does not require Congressional approval, meaning it could actually come to fruition, realigns incentives in the interest of individual investors by requiring retirement financial advisers to follow an established standard (a “fudiciary standard”) to act in their clients’ interest.

....The new fiduciary standard should block what honest brokers call “over-managing:” unnecessary rollovers, churning (over-active buying and selling that generates brokers’ fees at the expense of returns), and the pushing of expensive and risky products like variable annuities.

All of which turns out to be extremely costly to retirees....Conflicted advice reduces returns by about 1 percent per year, such that a poorly advised saver might end up with a 5 percent vs. a 6 percent return. They multiply that 1 percent by the $1.7 trillion of IRA assets “invested in products that generally provide payments that generate conflicts of interest” and conclude that the “the aggregate annual cost of conflicted advice is about $17 billion each year.”

According to Bernstein, a White House study suggests that this difference between 5 and 6 percent returns can amount to five years of retirement savings under plausible assumptions. That's a lot.

Needless to say, the financial industry is strongly opposed to this rule change, and I think we can safely assume that this means Fox News will be raising the alarums too. Their argument, apparently, is that if they're prohibited from giving small clients bad advice, it just won't be worth it to bother with small clients at all. Maybe so. But as Bernstein says, if that's really the case then "maybe there’s a hitch in your business model."

This has the potential to be an interesting campaign issue. Most Democrats, even those with close ties to the financial industry (*cough* Hillary *cough*) should have no trouble supporting this rule change. That's a slam dunk winner with retirees and most of the middle class. Republicans will have a harder time. After all, this represents regulation, and Republicans oppose regulation. They especially oppose financial regulation, as they've proven by their relentless efforts to roll back even the modest Dodd-Frank regulation adopted after the financial crash.

So what will they do? Stick to their principles and oppose the new regs? That will sure provide Democrats with an easy sound bite. Jeb Bush opposes a rule that prevents brokers from deliberately giving you bad retirement advice. I don't think I'd like to be the candidate who has to answer for that.

Is It Fair to Keep Peppering Scott Walker With Gotcha Questions?

| Mon Feb. 23, 2015 12:56 PM EST

Lately Scott Walker has been asked:

  • Whether he agrees with Rudy Giuliani's comment that President Obama doesn't love America.
  • Whether he believes in evolution.
  • Whether he believes that Obama is a Christian.

Is this fair? Why is Walker being peppered with gotcha questions like this? Are Democrats getting the same treatment?

There are no Democrats running for president yet, so it's hard to say what kind of questions they're going to be asked. But if Hillary Clinton attends a fundraising dinner where, say, Michael Moore suggests that Dick Cheney should be tried as a war criminal, I'm pretty sure Hillary will be asked if she agrees. And asked and asked and asked.

As for the other stuff Walker is being asked about—evolution, climate change, Obama's religion, etc.—there really is a good reason for getting someone like Walker on the record. He's basically a tea party guy who's trying to appear more mainstream than the other tea party guys, and everyone knows that there are certain issues that are tea party hot buttons. So you have to ask about them to take the measure of the man. Sure, they're gotcha questions, but they have a legitimate purpose: to find out if Walker is a pure tea party creature or not. That's a matter of real public interest.

Conservatives are complaining that Walker is facing a double standard. Maybe. We'll find out when Hillary and the rest of the Democratic field start campaigning in earnest. But I'm curious. What kinds of similar questions would be gotchas for Democrats? Drivers licenses for undocumented workers? Support for single-payer healthcare? Those aren't really the same, but I can't come up with anything that is. It needs to be something that's either conspiracy-theorish or else something where the liberal base conflicts with the scientific consensus, and I'm not sure what that is. GMO foods? Heritability of IQ? Whether George Bush stole the 2004 election by tampering with voting machines? I'm stretching here, but that's because nothing really comes to mind.

Help me out. What kinds of Scott-Walkerish gotcha questions should reporters be saving up for Hillary?