Hacker News Reader

Comment by jandrewrogers | Hacker News Reader

Actually, the studies adjust the incomes for purchasing power parity, so it is not just nominal income. Costs of living and similar are built into the relative income quintiles and deciles when comparing between countries.

The US also has a wider income range because of the enormous differences in cost of living within the country itself, in large part due to its size, which skews the result. The top quintile in one region of the US may be the middle quintile in another.

View on HN

Yes, after reading the article I'm not sure how they correct for the stark differences between places. Are the quintiles done locally or nationally? If they're done locally, then people who migrate are a problem for the statistics: it's very easy to move from bottom quintile to top quintile, just move from Palo Alto to rural Oklahoma. Or by excluding people who migrate, perhaps there are skewed effects.

But if the quintiles are national quintiles, not local quintiles, variation due to location still throws things off. The group of people in the 3rd national quintile living in Silicon Valley probably have quite different statistics than the group of folks in the 3rd national quintile living in a mid-size town.

One of the criticisms of the study is just this point. A lot of the "upward mobility" is people moving to more expensive areas where they get somewhat higher pay but the cost of living is much higher. They may well be worse off, but in $USD terms they have moved up.

Given that physical mobility is down in the US, this means that income mobility for reasons other than physical mobility must have increased.

http://www.nytimes.com/2013/12/15/magazine/why-are-so-many-a...

Costs of living and similar are built into the relative income quintiles

One way to break a correlation that uses an inflation index is to look at or carve out asset inflation. Once you correct for asset inflation (which is NPV'd income), only then will you have properly adjusted income measures.

Reply on news.ycombinator.com