[crpatino]

That is a falsifiable statement, you know.

Is there any statistic to show what percentage of today's top 1% earners where born to a bottom 50% family from the previous generation, and viceversa? If less than what you would expect from two non-correlated random variables, this suggest where you start in life has an effect of how far you can go. This does not invalidate your hypothesis, but may suggest that upwards mobility takes more than one generation to lift people from poverty to wealth.

Further more, we can do the same analysis to figure out how many people from bottom 50% families grow to reach 75% percentile or above. I do not know what it would be, but if much lower than expected, that would suggest that upwards mobility is quite limited, invalidating your hypothesis.

[lars]

You seem to be talking about average wealth and social mobility. This is different from inequality, which is about the distance from the bottom to the top.

Some of these results are counter intuitive. For example: There's data to show that the top quartile of earners in low inequality countries live longer than their top quartile counterparts in high inequality countries, even if the high inequality counterparts earn much more both in absolute and relative terms.

Edit: Since this is getting downvoted, here's a talk that spends 17 minutes listing studies showing that this is in fact about inequality, not average income. https://www.youtube.com/watch?v=cZ7LzE3u7Bw

[crpatino]

Right, I have seen that research before, and I don't think you should be getting down-votes either.

The relevance of social mobility in the discussion was because the GP assumes the correlation that the problems related to inequality/poverty is the result of a third factor that is cause of the other to: personal choices.

My position is that if this is the case, we should be observing strong social mobility in both ways. Surely there are a number of "deserving poor" that started life in a disadvantaged situation but are able to lift themselves out of that by intelligent choices and hard work, as well as there are privileged folks who make enough ill choices and land themselves in trouble.

IF this is not the case, we should adjust the level of agency that people are capable. To what degree is it personal choice able to influence the well being we enjoy in life.

[gwern]

> If less than what you would expect from two non-correlated random variables, this suggest where you start in life has an effect of how far you can go.

Where you start in life is not independent of innate traits.

[crpatino]

Agreed. But if that's the case, how (or to what degree) can inequality be the consequence of "one's own chosen behavior".

[gwern]

If, for example, things like impulsiveness and self-discipline (cough Conscientiousness cough) have high degrees of heritability, it can be simultaneously true that one's birth place correlates with one's destination and it also be the consequence of one's own chosen behavior. (And to the extent that the environment is equalized and optimal, genetic factors will matter more, not less.)