Inequality in Context

Via Brad Plumer, Jubin Zelveh reportson the development of new methods to measure inequality based on historical data.

His table is reproduced here and you can see that, as Jonathan Cohn could have told you, Denmark is awesome. I'm a bit surprised to see China in the late-nineteenth century and the Kingdom of Naples in the early nineteenth century come out as relatively egalitarian. My understanding had been that primarily agricultural societies are almost always super unequal since wealth (i.e. land) tends to be more unequally distributed than income, but iin societies like that a very large share of income goes to landowners as such. But perhaps not. He also says we have a ways to go in terms of upward redistribution of wealth:

It turns out that the typical modern nation has extracted about 33% of the available inequality (for the U.S. it's about 41%, for China it's 47%) while the researchers' sample of past societies squeezed out almost all of the available inequality.

Something to look forward to?

Comments

A single numerical measure can represent societal inequality and make the first century Roman Empire comparable to the US and modern Namibia?

I'm sure slaves in Rome and Byzantium would have been gratified to know that their society was more equal than our modern democracy.

More calculus please.

oy.

Posted by southpaw | November 3, 2007 1:22 PM

let's just pretend like this post never happened. move along, folks.

Posted by expat lumberjack | November 3, 2007 1:53 PM

Who cares if we're less equal than Rome, we have t.v.

Posted by Bill | November 3, 2007 2:25 PM

Sorry Matt, the article (and the study) you linked takes statistical illiteracy (innumeracy?) to realms previously unimagined. It's not even wrong.

If you have a sufficiently large and uniformly impoverished peasant class in a nation with minimal economic infrastructure (as in most feudal societies), and a sufficiently small aristocracy, your Gini coefficient will be quite low. That's why 1880 AD China and 1000 AD Byzantium score so high.

It's just a weakness of the measurement; it doesn't tell us anything useful.

Posted by heedless | November 3, 2007 2:31 PM

It's just a weakness of the measurement; it doesn't tell us anything useful.

Sure it does. It reaffirms my suspicions about Namibia.

Posted by Bill | November 3, 2007 2:35 PM

The graph seems suspect. The western Roman empire was a terrible place to live -- after the empire fell the average height went up in France (Francia? I'm feeling too lazy to look up the right name for 600 CE). It was measured from skeletons from graves before and after Roman rule.

I do remember from my Chinese history course that the life of a peasant in China was, on average, better than in medieval Europe. I don't have any links to support that though.

Posted by stm177 | November 3, 2007 2:53 PM

Cliometrics!

I'd note that MY's assumption that _all_ agricultural societies tend towards greater inequality is a perfect illustration of the flattening tendencies of the social sciences; that's why we have History, for nuance and particularity and all that junk.

Posted by Ben Cronin | November 3, 2007 3:36 PM

Also, a couple weeks ago Matt had a link to England vs. stone age nutrition, and life expectancy.

I posted it here at a history forum, and one of the history geeks pointed out that the life table can be misleading to amateurs. Life expectancy in such a table is the time left to live from the age listed. So, if it says birth, and a life expectancy of 34 years, then an infant has 34 years left to live.

If it says "at 20 years", and 32 years life expectancy, you add the two numbers like so: 20+32=52.

This seems like a good place to post this kind of thing, since I remember people wondering what was going on if a 20 year old would only live to age 32. There was some speculation on the Napoleonic wars and such. I'd think people interesting in this post of Matt's would correlate with the earlier post of his.

Posted by stm177 | November 3, 2007 4:39 PM

Re: The western Roman empire was a terrible place to live -- after the empire fell the average height went up in France

Largely because the population crashed like a lead balloon after a series of natural disasters in the 6th century. Fewer people=higher standard of living as the survivors divvy up the wealth. Same thing happened in the 15th century after the Black Death brought the numbers down for a while.

Re: So, if it says birth, and a life expectancy of 34 years, then an infant has 34 years left to live.

No, that's a bad way of stating it. What it means is that an infant has a 50 per cent chace of dying before reaching 34 and a 50% chance of dying afterward. Pre-modern (well, pre-20th century) life expectancies are always skewed downward by high childhood mortality rates. In some eras 2/3 of all children born did not make it to adulthood-- and the upper classes were not free of this hard fact of life either, if you read the annals of any royal family in the past.

Posted by JonF | November 3, 2007 5:15 PM

So, England in 1688, the year of the Glorious Revolution, the year after Newton's book, a couple of years before Locke's works, was much more unequal than decadent, immobile China in 1880. And England in 1801, as the Industrial Revolution was picking up steam, was much more unequal than the Kingdom of Naples in 1811.

Well, I'm convinced! Sign me up for your crusade to turn America into China and/or Naples at the low points in their history.

Posted by Steve Sailer | November 3, 2007 7:16 PM

Time for my usual bitch - why, why, why is it so &^%$ difficult to put a freakin' label on the Y axis? This graph is meaningless without some kind of explanation of what these numbers represent. If you didn't already have some a priori knowledge that Denmark was more egalitarian than Namibia, you'd be completely in the dark about what this graph meant. And as it is, you don't know much.

Posted by Sean Peters | November 3, 2007 8:29 PM

I think the Nays have it.

Will Yglesias defend this post?

Posted by southpaw | November 3, 2007 8:37 PM

Ugh... Matt, just stay away from magical new statistical measures that purport to say something non-intuitive, because usually they're terribly misleading and apparently whatever you were studying at Harvard didn't include much in the way of statistics. Stick to basketball and letting us know what's going on inside the beltway.

Posted by Adam Villani | November 4, 2007 5:09 AM

Actually, it's perfectly plausible for China in 1880 to have a more even distribution of wealth than, say, England in 1688.

The wealthier a society, the more unequally income can be distributed without bringing on a total societal collapse or a forced redistribution of income. Industrial and high-trade (commercial) societies can sustain a much higher level of income inequality than agricultural ones because they have a much larger pie to divide. What is left over after the majority end up with just enough to survive is much larger for such societies.

They don't HAVE to have higher income inequality, but they can sustain a higher level. And since industrialization and trade, unless counter-balanced by things like unions or social welfare measures, tends to concentrate wealth in a few hands, there's a strong tendency for it to happen.

England in 1688 was a time of increasing income inequality. More and more farmers were losing their farms and being reduced to poverty as the major land holders consolidated their landholdings. Merchants grew in wealth from growing trade, but their employees largely got paid very little. Merchants and aristocrats and gentry land holders grew in wealth; everyone else was going down hill. By the 18th century, 33% of the English population was poor. (By contrast, in largely agricultural America, only 10% of the population was poor, as it was much easier to own your own farm and be self-sufficient.)

China, in 1880, followed the system of small family farming which it had followed since the time of the Song Dynasty (900s-1200s AD). Most of the ruling class lived comfortably, but didn't have ungodly amounts of wealth. Only the Imperial family had huge amounts of wealth. That's not to say that the Confucian-trained aristocracy was starving, but they mostly lived a lot closer to the standard of living of the peasantry than did 18th century English aristocrats.

However, because China was not very wealthy relative to its population size, the position of the Chinese peasantry wasn't too good, despite a less unequal distribution of wealth.

Posted by John Biles | November 4, 2007 1:53 PM