« Satire - A Dessert Best Served Cold | Main | Down With the Pasty Sandwiches of Privilege! »

November 21, 2013

Once Again, Science Has Spoken....

...and its voice is Imperial:

...why is it important that we have a multitude of desperate law school graduates and many more politically ambitious rich than 30 years ago?

Past waves of political instability, such as the civil wars of the late Roman Republic, the French Wars of Religion and the American Civil War, had many interlinking causes and circumstances unique to their age. But a common thread in the eras we studied was elite overproduction. The other two important elements were stagnating and declining living standards of the general population and increasing indebtedness of the state.

Elite overproduction generally leads to more intra-elite competition that gradually undermines the spirit of cooperation, which is followed by ideological polarization and fragmentation of the political class. This happens because the more contenders there are, the more of them end up on the losing side. A large class of disgruntled elite-wannabes, often well-educated and highly capable, has been denied access to elite positions. Consider the Antebellum U.S.

From 1830 to 1860 the number of New Yorkers and Bostonians with fortunes of at least $100,000 (they would be multimillionaires today) increased fivefold. Many of these new rich (or their sons) had political ambitions. But the government, especially the presidency, Senate and Supreme Court, was dominated by the Southern elites. As many Northerners became frustrated and embittered, the Southerners also felt the pressure and became increasingly defensive.
Slavery had been a divisive force since the inception of the Republic. For 70 years, the elites always managed to find a compromise. During the 1850s, however, intra-elite cooperation unraveled. On several occasions Congress was on the brink of a general shootout. (As one senator noted about his “armed and dangerous” colleagues, “The only persons who do not have a revolver and a knife are those who have two revolvers.”)

Hmm. Hard to argue with that. We know this is Scientific, because there's a diagram. With lots of ... arrows. And cartoon figures.


With the End of Times so near at hand, can baby killing and puppy bouncing be far behind? Science asks, you decide:

Abstract: Background: Infant mortality rates in the US exceed those in all other developed countries and in many less developed countries, suggesting political factors may contribute.

Methods: Annual time series on overall, white and black infant mortality rates in the US were analyzed over the 1965-2010 time period to ascertain whether infant mortality rates varied across presidential administrations. Data were de-trended using cubic splines and analyzed using both graphical and time series regression methods.

Results: Across all nine presidential administrations, infant mortality rates were below trend when the President was a Democrat and above trend when the President was a Republican. This was true for overall, neonatal, and postneonatal mortality, with effects larger for postneonatal compared to neonatal mortality rates. Regression estimates show that, relative to trend, Republican administrations were characterized by infant mortality rates that were, on average, three percent higher than Democratic administrations. In proportional terms, effect size is similar for US whites and blacks. US black rates are more than twice as high as white, implying substantially larger absolute effects for blacks.

Conclusions: We found a robust, quantitatively important association between net of trend US infant mortality rates and the party affiliation of the president.

Meh. Needs arrows to tighten up the causal connection, wethinks.

OH THANK GOD. Andrew Sullivan weighs in: Grim. Challenging. Persuasive. Crude. Powerful. Elite Fratricide.

Posted by Cassandra at November 21, 2013 01:31 PM

Trackback Pings

TrackBack URL for this entry:


I would appreciate your putting more distance between his name and mine. :)

Posted by: Grim at November 21, 2013 06:50 PM

FYI, a primary reason America has higher infant mortality rates than other developed nations is that we count them much differently. Here, we count nearly all late term pregnancies (sadly, abortions are excluded) that do not produce a living child as a mortality.
Many countries exclude some infant deaths (stillborn as an example), or fail to count all of the deaths of young infants for varying reasons.
It's not an apple to apple comparison.

Best Regards,

Posted by: CAPT Mike at November 21, 2013 07:32 PM

Once Again, Science Has Spoken....and its voice is Imperial, "Ekki-Ekki-Ekki-Ekki-PTANG. Zoom-Boing. Z'nourrwringmm."

Posted by: DL Sly at November 21, 2013 07:49 PM

I'm still waiting to hear what Yu-Ain thinks of the infant mortality abstract. I particularly enjoyed this part:

Regression estimates show that, relative to trend, Republican administrations were characterized by infant mortality rates that were, on average, three percent higher than Democratic administrations. In proportional terms, effect size is similar for US whites and blacks. US black rates are more than twice as high as white, implying substantially larger absolute effects for blacks.

Posted by: Cass at November 21, 2013 09:33 PM

Here's an interesting fact (or maybe factoid -- I haven't checked it personally): According to a post at Political Calculations, among individuals there has been no increase in income inequality.

Let me repeat, no increase. None, Nada, Zero.

The increase in family income inequality can be ascribed to relative changes in family structure. Or to put it more directly, now, rich folks are much more likely to marry and stay married than poor folks.

If that argument is correct -- and I think it is, though I haven't, as I said, done a step-by-step check of the data and the reasoning -- then we probably want to look at our inequality problems differently than Turchin suggests.

Posted by: Jim Miller at November 21, 2013 10:51 PM

Yes, but where is the arboreal gap in that graphic? Obviously someone did not get the scientific consensus e-mail.

As to the "boom" part, it's nothing more than informed citizens laughing at the politicians.

"We're serious, and we know what we are doing."

Snicker, sure you are, can you change a flat tire?

"No we mean it, this is going to be great."

Guffaw, can you do basic geometry, what are parallel lines?

"Look we can create jobs and grow the economy."

You're killing me, describe a job.

Posted by: Allen at November 22, 2013 05:01 AM

The abstract on the infant mortatlity study doesn't particularly set off any red flags.

As Capt Mike noted, Infant mortality rates are not comparable across national lines as some countries report deaths within the first week of birth as "non-births" and so adds neither to the numerator nor denominator.

But just because their justification for thinking that politics are a factor are bogus doesn't mean that what they find isn't there.

I also agree that the data needs to be "de-trended". That is, that infant mortality rates have decreased over time simply because of medical advances. Babies that were beyond saving in 1965 can now be saved on an almost routine basis. They are correct that this should not be attributed to the political climate. I don't know that cubic splines would have been the tool I'd use to do that, but that is neither here nor there.

Of more interest would be in how these splines* are choosen. If the joining nodes are placed one year into each presidents term (when his policies would might start to have an effect) then you would attributing a good portion of that president's effect not to the policy but to the trend. They claim the model is robust (that is, minor changes don't effect model outcomes very much), but I'd be currious to see how robust it is to node changes. Because if the model doesn't change because of the spline scheme, why did they need it?

Which isn't to say that it's wrong. It's just something that is not (and would not be) in the abstract.

I also wish the abstract laid out more about what other factors were investigated for model inclusion. Economic factors (income/insurance), health factors (smoking/drinking/prenatal care), age of mother, length of practice for the MD, etc would all have effects that need to be accounted for.

Even in the political sphere, why choose the President? Why not the majority control of the Senate and/or House? The President has the bully pulpit, but congress determines policy, the POTUS just implements it.

I'd also be interested in the time lag. As we've seen, policy doesn't get implemented over night. ObamaCare didn't get passed on Day 1 of Obama's term and it's legal implementation didn't happen until his second term, (and practical implementation is still up in the air).

The last issue isn't a statistical one, but one of priorities. They found a consistent, and huge factor of race (a 100% increase for blacks) but what they focus on is the 3% increase/decrease by political party of POTUS. Talk about not seeing the forest for the trees.

*A spline is an "if-then" type of regression: if the data is in one range then you use one regression model, and if the data is in a different range then you use a different regression model, etc. You have to be careful so that the end points of each spline connect to the next one and that the join is smooth and doesn't have a corner.

Posted by: Yu-Ain Gonnano at November 22, 2013 09:58 AM

"The optimal moment for successful societies is when the middle class dominates, where political institutions reflect a mass interest in governing the society well, because everyone feels they have a stake (so more people than usual want and need collective success), and because they share some basic commonalities in experience, and so can find a way to compromise." Later, he describes "compromise" as "sharing the wealth."

I was going to rob a bank of $10K once. I offered to compromise and take $5K. The bank was all "You're stealing our money." I just wanted a stake in it.

Posted by: Texan99 at November 22, 2013 10:05 AM

the join is smooth and doesn't have a corner.

I tank this back. It's not strictly true that you can't have a corner, but my opinion is that you better have a darned good reason for thinking a corner really does exist to allow one.

Posted by: Yu-Ain Gonnano at November 22, 2013 10:08 AM

I just want to point out that unemployed lawyers leads almost directly to tornados, which we all know is code for global warming. If you care about the environment, you will not cause lawyers to become unemployed.

Posted by: Manbearpig at November 22, 2013 02:26 PM

I'm not entirely convinced that the incomprehensibility of ObamaCare wasn't an attempt to create new jobs for lawyers.

Posted by: Yu-Ain Gonnano at November 22, 2013 02:52 PM

Hell's bells, YAG may be more literal than I.

Posted by: CAPT Mike at November 23, 2013 01:16 AM

Well? She *asked*. :-)

Posted by: Yu-Ain Gonnano, Your Friendly Neighborhood Statistician at November 25, 2013 09:20 AM