Mathematician Gary Cornell has a theory I like about perceptions of intergenerational mobility in different countries and an empirical question the theory depends on.  Please send replies to him at gary@thecornells.com . I will find out from him what answers he got.  Here is his email I reprint with his permission with his theory and question:

Hi Miles, thanks for the post.

I wonder if you know anything about the following conjecture of mine:

The background is that I’ve long puzzled about people believe that America is the land of opportunity even though intergenerational mobility  is actually lower here than in many other countries. I  have a conjecture  that seems to have some  evidence to sustain it. (Basically I scanned the equivalent of the Forbes rich lists here and in other countries).

Namely:  While moving from upwards from one quintile to another  is less likely in the united states, moving  to the top .01% of wealth (assets of about $30 million roughly) is far more likely in the United States than in other countries or even moving into the top .1% (about 7 million in assets)  (For these people think of the first 1,000 people hired at Google or the first few hundred at Facebook  for example… )  In other countries I am pretty sure that the probability  that someone born  middle class or even upper middle class can become really really rich is a small fraction of what it is here.  So more precisely my conjecture is that, for example, because the ratio:

P(getting rich in the US)/P(getting rich in Europe)  

is so large people (and the media) tend to believe and the media tends to report that America is a land of opportunity compared to other countries and to forget that

P(getting really  rich in the US)

is almost vanishingly small in absolute terms… (If you look at the equivalent of the Forbes list in other countries it sure looks like the american list has fewer examples of people getting their place through the lucky sperm derby than in England, France. Or Germany.) Do you know of any papers on this?