Why Observational Epidmeiology is frustrating

Andrew Gelman has a post up on the history of cigarette smoking research, based on a book he was reading a while back.  It's pretty interesting but what really caught my eye was this comment:

Vague statistical inference can not possibly establish such a causal link. Even valid associative inference should establish a 50-100% correlation between smoking and cancer, but it does not even come close. Most people who smoke don’t get lung cancer, and at least 10% of Americans who do get lung cancer- do not smoke. There are also huge international/ethnic variations among smokers and cancer rates. There is currently no proof whatsoever for the alleged smoking-cancer causal link. None. Smoking is a disgusting and silly habit. But all that one can now objectively say is that it is a risk factor for cancer and increases the incidence of lung cancer.
 

And this was not the only person in the comments who was casting doubt on this association.  As an epidemiologist, I want to scream.  If people will not believe this evidence then they really will not believe any level of evidence for observational epidemiology.  We have cohort studies going back 50 or more years (Richard Doll has one). Even better, the members of this cohort did not initially know that smoking was harmful (and I recall that the original hypothesis was automobile fumes and not smoking, although my memory may be failing me here).  So we don't even suspect a healthy abstainer effect. 

The requirement for a 50 to 100 correlation seems to ask for smoking to be directly causal of lung cancer instead of increasing the underlying risk of lung cancer.  Consider skiing and broken legs.  Not all broken legs are due to skiing and many people ski and do not break a leg.  But there is no question that skiing is a risk factor for broken legs.  Another good example is collapsed disks in the back.  If you are working with a veterans population, the first question you ask when you see a compression fracture in the spinal cord is "were you a paratrooper?".  Not all paratroopers have compression fractures and not all compression fractures are due to jumping out of airplanes, but it is a pretty direct link to increased risk. 

There is a libertarian line of defense here: people ski because they value the enjoyment of skiing more than the risk of a broken limb.  I am not always delighted by it, but it is at least an arguable position.  But directly denying the link between smoking and lung cancer seems to be setting a very aggressive standard of proof. 

Felix Salmon is back: California pension edition

Felix has a very nice post on the new retirement program that California is considering.  I think it is a very good idea and the state running it is a major plus.  It's astounding but true that the very best instruments that I can find in the private sector (currently thinking Vanguard here) seem to run neck and neck with large retirement plans in terms of both fees and returns.

I am a little less thrilled by investment guarantees, unless they are really well adjusted for inflation.  But, conditional on that, this would also smooth out returns in a very nice way.  Sure, it has the government investing in the markets.  But the only other alternative is paygo, which has come under a lot of question as to the willingness of taxpayers to honor previous promises.

I agree that this is an exciting development.

This will sound like a snarky question, but it's not

Via a long chain starting here, came across the following article on Romney's missionary service:

Republican presidential candidate Mitt Romney could be excused for having flashbacks to the 1960s when he went door to door in Berlin, New Hampshire, on Thursday.

The former Massachusetts governor worked in France as a Mormon missionary from 1966 to 1968, one of the church’s thousands of earnest young men (mostly) who knock on doors and proselytize. At that point Romney had plenty of doors slammed in his face, but on Thursday, not so much.

“This is a lot easier,” Romney quipped to Reuters. “People speak English. They wish you Merry Christmas. They don’t think you’re a salesman. People used to come to the door [in France] and wag their fingers: ‘No, I don’t want anything.’”

Many French people at the time were “not happy to see Americans, because we were in Vietnam at the time. That was tough,” he added.

I realize the war was unpopular in most of Europe, but I'm a bit surprised to hear it was unpopular in France. given their relationship with the country and the conflict. I was under the impression (backed up by a quick visit to Wikipedia) that the US was, in a sense, picking up where the French had left off. I even seem to recall Eisenhower saying that we should be careful about getting involved with Vietnam because it was a continuation of a pro-colonial war (am I remembering this right?).

How strong was the anti-war feeling in France in the late Sixties?

This is getting old,,

I've been getting this this screen frequently and seemingly randomly (it seems to have a special fondness for Economist's View). Is this a Chrome thing? A Microsoft thing? A techie conspiracy to get back at me for that whole ddulite business?

What's wrong with the blogosphere?

After the GOP convention, are they actually letting this pop culture allusion go by without pounding it into the ground?

Smart meters, incentives and a chance to use "extruded" in a post

Here's a product idea for you, an old notion repackaged for a new problem. It's a piece of plastic in the shape of an extruded upside-down T filled with a saline or alcohol solution that freezes exactly at a certain temperature (let's say 20 degrees). During the summer, you put the T into your freezer (I'm assuming the freezer is on the top) and set the thermostat timer (did I mention you need a thermostat timer?) to turn the freezer down to ten degrees (I'm using round numbers here. Actual values would certainly be different) from one a.m. to five a.m. then up to just under twenty degrees from five to ten then up to twenty-five from ten to four then just over twenty from four till one the next morning.

Obviously, I didn't put a lot of research into freezers before coming up with these temperatures, but I think you catch the basic idea: have the freezer work hardest in the middle of the night when electricity is cheap and the house is cool and do as little as possible when the grid is overloaded.

There's nothing original here. Old-fashioned ice-cream makers use the same principle. Nor is there anything unusual; without breaking a mental sweat, you could probably come up with a dozen comparable (or better) ideas for smoothing out the power demand curve.

By making more efficient use of our electrical resources, these ideas would pay for themselves in... well, never. You pay the same for the electricity you consume at two in the morning as you do for what you use just before a brown-out. That's the problem. There are lots of things we could do to smooth out our power demand curve but under the current system (pardon the pun) there's no incentive.

There's a notable precedent we can refer to. The current generation of of automobiles feature an array of innovative energy efficient technologies. The feature that impressed me most as a driver was the continuously variable transmission. An engineer pointed out that many of those energy saving technologies, including the CVT, had actually been around for decades, but had only come into wide use because regulations and/or high prices (think Europe) and consumer demand had forced engineers to find ways to squeeze every possible mile out of a gallon of gas. We can argue about what kind of incentive worked best to spur innovation, but no one can argue that incentives didn't work.

Of course, we do have the technology for variable demand-based pricing, and we have a line of incredibly cheap credit to pay for implementing it. What we don't have is a real interest in solving problems.

And a lack of interest in solutions is a difficult problem to solve.

is $152B < $186B

Opinions differ.

Urban Libertarians

This comment thread go me thinking about something that's been bouncing around my head for a while now. Living in a city or a suburb puts severe limitations on what it means to own a house. (the case is even more extreme with condos but I don't really consider that ownership. Those are just apartments with mortgages.) Between HOAs and city ordinances, you have remarkably little freedom to do what you want with your own property.

That's just the start. Life in the city is constrained by a dense array of rules, because

1. Even as a homeowner, you spend a great deal of time in property you don't own;

2.  Cities encourage specialization and a correspondingly high degree of interdependence;

3. It takes a lot of rules to allow this many people to live in this close a quarters;

4. If my freedom to move my arm stops at the closest person's nose, crowds have to severely limit my freedom.

I'm a city dweller, but if I gave into my libertarian tendencies, the only place I could stand living is back in the country. I'd get a place in the lower Ozarks off highway 7 at the end of a dirt road where a man can do what he damned well pleases. I'm not planning on moving any time soon but there are times the notion hits me (besides, it bothers me that I've forgotten that distinct smell of a dirt road).

Which is why libertarian urbanists like Edward Glaeser confuse me so. If the really value liberty so highly, why do they embrace the most rule dependent of lifestyles?

Stories worth catching (Public Radio Edition)

From All Things Considered:

"One of the amazing things about the Denisovans is we still don't know what kinds of artifacts they left behind, what kinds of archaeology. We really only know them from their DNA," he says. "So it's really a genome in search of an archaeology."

From Marketplace:

Well, Cornell University professor David Just tried a little experiment on a few hundred preschoolers. He put cookies and apples in front of them at lunch time, and then got the number of kids who put an apple on their lunch tray to increase by 70 percent.

And this excellent piece of financial journalism from This American Life. (only complaint -- a "Measuring Life" would have been a much better title).

More on the challenges of modern media

More media issues by Jon Chait:

In the campaign press corps, Romney’s brazen decision to not merely indulge in puffery or invective or half-truths — as all campaigns, Obama’s included, have done — but to base his entire message on straight-up lies has prompted some journalistic soul-searching about the role of the campaign press. Reporters have come to outsource the role of evaluating the truth of candidates’ claims to “fact-checkers.” This allows the reporters to avoid directly calling a candidate a liar, but instead to point out that some third party, the fact-checkers, have called them liars. The utility of this arrangement has brushed up against its natural limit, a fact that was brutally exposed when Romney pollster Neil Newhouse asserted, “We’re not going to let our campaign be dictated by fact checkers.”
 

This really does make the traditional "the truth is in the middle" approach to reporting difficult.  At some point one presumes that a healthy press simply needs to start calling out actual mis-truths if they want to avoid being unpaid PR flacks.  At some point the role of reporting has to involve at least a nod to truthful discourse. 

Dealing with homeowners' associations has got to be hell for libertarians

I'm serious. Is there any other institution so focused on making and enforcing trivial and often idiotic rules?

From Marketplace:

If you have trouble keeping your lawn green in the summer time, here's an incentive: a $200 dollar ticket. That's what some homeowners in one Denver development are facing for brown patches in their yards. Yes, in the same state that just experienced a rash of massive wildfires.

You need to listen to this story on the threat of a global food crisis

No, I really mean you have to listen to it if you want to follow the story -- the text version is incredibly skimpy.



 

More stories on infrastructure

At some point one assumes that this system must be terribly broken:

Littlefield also argues that judges in New York routinely side with contractors in disputes with the Metropolitan Transportation Authority. “In the private sector, if you rob your customer, you will suffer a hit to your reputation and possible losses in the courts,” he said in an interview. “Not so if you rob an agency like the MTA. Then it’s all rights and no responsibilities.”

The MTA must continue to award contracts to the lowest- price bidder, and without the ability to hold bad contractors accountable, Littlefield said, the agency turns to “writing longer and longer and longer contracts, expressly prohibiting every way it has been ripped off in the past.” The byzantine contracts that come out of this process drive entrants away, limiting competition and pushing up costs.

The inability to reward contracts based on reputation is a huge problem.  It means that there is a huge amount of  benefit to thinking of a way to rip off the local government and no real consequences to doing so.  Market forces don't work because the need to give the award to the lowest bidder, regardless of reputation, makes it impossible to stop doing business with a bad vendor so long as they have the lowest price. 

Imagine if you had to buy a hamburger this way.  In the real world we know that there is a quality difference between hamburgers as well as a substantial price difference.  But if you always had to buy the lowest cost hamburger unless you could rule out a feature of the burger in advance, you can imagine how hard it would be to get the hamburger that you wanted and how easy it would be for the vendor to play games.

It's not a good system and I can see no reason why we should not look to reform it. 

Seeing things differently

This recent bit of advice for Mitt Romney from Peggy Noonan (by way of Brad DeLong) brought home how perceptions can differ from party to party:

[Romney] must use humor, for three reasons. One is that wit breaks through and sharpens all points. Another is that it is natural to him. Before the voting in Iowa, he wryly told a friend that the caucuses were like the LaBrea Tar Pits: “No one comes out the way they went in.” On a conference call recently, he asked a question of his staff. No one answered. Mr. Romney waited. “Bueller? Bueller?” he said, in a perfect imitation of Ben Stein.

Third, President Obama can’t stand to be made fun of. His pride won’t allow it, his amour propre cannot countenance a joke at his own expense. If Mr. Romney lands a few very funny lines about the president’s leadership, Mr. Obama will freak out. That would be fun, wouldn’t it?

If I were asked to rank recent Republican nominees on being naturally funny (thus leaving the partisan question out, since perception of humor is likely to be influenced how much we agree with someone), I'd put Reagan and McCain in the first class, Dole and George W. in the next, H.W. in the third. Romney would rank the lowest.

As for Obama's tendency to "freak out," I tend to see just the opposite, someone who keeps his cool and is exceptionally difficult to fluster.

These are, of course, subjective qualities, but it seems like the gap between the way different sides see these qualities has gotten almost inexplicably large.

(Illustrations courtesy of Gustave Verbeek.

Robert H. Frank comes out strong for the carbon tax

While we're on the subject, check out this op-ed in the New York Times:

Mitt Romney, for his part, has been equivocal about whether rising temperatures are caused by human action. But he has been adamant that uncertainty about climate change rules out policy intervention. “What I’m not willing to do,” he told an audience in New Hampshire last summer, “is spend trillions of dollars on something I don’t know the answer to.”

...  

The good news is that we could insulate ourselves from catastrophic risk at relatively modest cost by enacting a steep carbon tax. Early studies by the Intergovernmental Panel on Climate Change estimated that a carbon tax of up to $80 per metric ton of emissions — a tax that might raise gasoline prices by 70 cents a gallon — would eventually result in climate stability. But because recent estimates about global warming have become more pessimistic, stabilization may require a much higher tax. How hard would it be to live with a tax of, say, $300 a ton?

If such a tax were phased in, the prices of goods would rise gradually in proportion to the amount of carbon dioxide their production or use entailed. The price of gasoline, for example, would slowly rise by somewhat less than $3 a gallon. Motorists in many countries already pay that much more than Americans do, and they seem to have adapted by driving substantially more efficient vehicles.

A carbon tax would also serve two other goals. First, it would help balance future budgets. Tens of millions of Americans are set to retire in the next decades, and, as a result, many budget experts agree that federal budgets simply can’t be balanced with spending cuts alone. We’ll also need substantial additional revenue, most of which could be generated by a carbon tax.

If new taxes are unavoidable, why not adopt ones that not only help balance the budget but also help make the economy more efficient? By reducing harmful emissions, a carbon tax fits that description.

A second benefit would occur if a carbon tax were approved today but phased in gradually, only after the economy had returned to full employment. High unemployment persists in part because businesses, sitting on mountains of cash, aren’t investing it because their current capacity already lets them produce more than people want to buy. News that a carbon tax was coming would create a stampede to develop energy-saving technologies. Hundreds of billions of dollars of private investment might be unleashed without adding a cent to the budget deficit.

SOME people argue that a carbon tax would do little good unless it were also adopted by China and other big polluters. It’s a fair point. But access to the American market is a potent bargaining chip. The United States could seek approval to tax imported goods in proportion to their carbon dioxide emissions if exporting countries failed to enact carbon taxes at home.

As mentioned before, many of the steps we can take are cheap and/or already necessary to maintain a competitive economy. Some even pay for themselves without taking climate change into account:

The railroad has long been reluctant to accept government investment in its infrastructure out of fear of public meddling, such as being compelled to run money-losing passenger trains. But now, like most of the industry, it has changed its mind, and it happily accepted Virginia’s offer last year to fund a small portion—$40 million—of the investment needed to get more freight traffic off I-81 and onto the Crescent Corridor. The railroad estimates that with an additional $2 billion in infrastructure investment, it could divert a million trucks off the road, which is currently carrying just under five million. State officials are thinking even bigger: a study sponsored by the Virginia DOT finds that a cumulative investment over ten to twelve years of less than $8 billion would divert 30 percent of the growing truck traffic on I-81 to rail. That would be far more bang for the state’s buck than the $11 billion it would take to add more lanes to the highway, especially since it would bring many other public benefits, from reduced highway accidents and lower repair costs to enormous improvements in fuel efficiency and pollution reduction. Today, a single train can move as many containers as 280 trucks while using one-third as much energy—and that’s before any improvements to rail infrastructure.

When someone like Mitt Romney talks about spending trillions, the phrase is only meaningful if we're talking about a difference of trillions between two courses of action. In the case of rail upgrades, spending a trillion might well mean hundreds of millions in savings compared to achieving similar capacity through other means.