5 ms·

Ugh. If you look at the sourced article[1] the effect is actually very small: "When regressing the pseudo-descendant’s earnings on pseudo-ancestor’s earnings,

Ugh. If you look at the sourced article[1] the effect is actually very small: "When regressing the pseudo-descendant’s earnings on pseudo-ancestor’s earnings, the results are surprising: the long-run earnings elasticity is positive, statistically significant, and equals about 0.04. Stated differently, being the descendants of the Bernardi family (at the 90th percentile of earnings distribution in 1427) instead of the Grasso family (10th percentile of the same distribution) would entail a 5% increase in earnings among current taxpayers (after adjusting for age and gender)" http://voxeu.org/article/what-s-your-surname-intergenerational-mobility-over-six-centuries http://voxeu.org/article/what-s-your-surname-intergeneration...

jerf 10y ago
This strikes me as proving an awfully small result too concretely. Across the several thousand ancestors you have going back 600-700 years (if not millions of ancestors; a rather conservative "one generation every 25 years" for 600 years yields 2^24-ish possible ancestors, ~16 million, though most people's family trees do not actually branch that much in the less mobile times of the past), the one from which your patrilineal name comes from is even 5% relevant to your life today? Is this one of those cases where hard science and math prove something very counterintuitive, or one of those cases where soft science and math gets popular because it was poked and prodded until it yields a counter-intuitive result that would get into the news?
- panglott 10y ago
  My thought was similar: how long ago was the most recent common ancestor of Florentines (barring immigration)?
shmageggy 10y ago
Thanks for pointing this out and citing the paper. When I saw the Andersson graph I thought, "oh hey, those two roughly Gaussian distributions are indeed a wee bit distinguishable", then I read > were drastically more likely to be in the Swedish financial elite than people with the surname Andersson. and I was like, "Drastically? Really?" Another day another shoddily reported science article.
- aab0 10y ago
  Two distinguishable Gaussian distributions will have very different behavior in the tails. So if you knew two groups had slightly different means in wealth, then it would not be surprising to discover that at a particular cutoff in the tail, one group will be much overrepresented. Thus if you look at Clark's paper http://faculty.econ.ucdavis.edu/faculty/gclark/papers/Sweden%202012%20AUG.pdf http://faculty.econ.ucdavis.edu/faculty/gclark/papers/Sweden... , you see that while incomes may only be 21% higher for descendants of noble surnames vs a normal surname, "we see that Noble Surnames are about 6 times more likely to be doctors or attorneys, and about 4 times more likely to be graduating an elite university in 1998-2012"; likewise, the mean increase in education is not that big, but the Royal Academy overrepresentation is something like 5x.
Houshalter 10y ago
That's not small at all. The expected effect size is 0% over so many generations and political upheavals. 5% is ridiculously high for 700 years!

jerf 10y ago

This strikes me as proving an awfully small result too concretely. Across the several thousand ancestors you have going back 600-700 years (if not millions of ancestors; a rather conservative "one generation every 25 years" for 600 years yields 2^24-ish possible ancestors, ~16 million, though most people's family trees do not actually branch that much in the less mobile times of the past), the one from which your patrilineal name comes from is even 5% relevant to your life today? Is this one of those cases where hard science and math prove something very counterintuitive, or one of those cases where soft science and math gets popular because it was poked and prodded until it yields a counter-intuitive result that would get into the news?

panglott 10y ago
My thought was similar: how long ago was the most recent common ancestor of Florentines (barring immigration)?

shmageggy 10y ago

Thanks for pointing this out and citing the paper. When I saw the Andersson graph I thought, "oh hey, those two roughly Gaussian distributions are indeed a wee bit distinguishable", then I read > were drastically more likely to be in the Swedish financial elite than people with the surname Andersson. and I was like, "Drastically? Really?" Another day another shoddily reported science article.

aab0 10y ago
Two distinguishable Gaussian distributions will have very different behavior in the tails. So if you knew two groups had slightly different means in wealth, then it would not be surprising to discover that at a particular cutoff in the tail, one group will be much overrepresented. Thus if you look at Clark's paper http://faculty.econ.ucdavis.edu/faculty/gclark/papers/Sweden%202012%20AUG.pdf http://faculty.econ.ucdavis.edu/faculty/gclark/papers/Sweden... , you see that while incomes may only be 21% higher for descendants of noble surnames vs a normal surname, "we see that Noble Surnames are about 6 times more likely to be doctors or attorneys, and about 4 times more likely to be graduating an elite university in 1998-2012"; likewise, the mean increase in education is not that big, but the Royal Academy overrepresentation is something like 5x.

Houshalter 10y ago

That's not small at all. The expected effect size is 0% over so many generations and political upheavals. 5% is ridiculously high for 700 years!