3 ms·
Citations? I think the top 1% is a much more stable group than this indicates. Top 10%, I would completely agree with here. And I fully accept I could be wro
by taeric 10y ago
Citations? I think the top 1% is a much more stable group than this indicates. Top 10%, I would completely agree with here.
And I fully accept I could be wrong. I'm currently googling a few things, to see if I am off here. I'm asking not to be a jerk, but this is put forward as fact here. So I'm assuming you have actual numbers to back this up.
- yummyfajitas 10y agoOnly 1/3 of people in the top 1% in 2005 would remain in the top 5 by 2009. See table 3: https://www.aeaweb.org/conference/2013/retrieve.php?pdfid=462 https://www.aeaweb.org/conference/2013/retrieve.php?pdfid=46... Among people with income > $1M, 50% had this income for only 1 year. https://taxfoundation.org/millionaire-status-fleeting/ https://taxfoundation.org/millionaire-status-fleeting/ One great way to have a lot of income is to sell a house that appreciated in value over many years. But it's hard to do that repeatedly.
- stuaxo 10y agoThis is so soon after the credit crunch of 2008 it can't be generalised, A data set ending 1 year after the great depression would probably look the same. It would be more interesting to take a longer term data, just like this study did. (Even 2005 to 2015 would be more realistic).
- yummyfajitas 10y agoDid you even read the paper? From 1991-2005, the odds of remaining in the top 1% for 5 years have ranged from 21% (in 1999) to 36% (in 1994). Intermediate results are more or less the same from 2005-2010.
- taeric 10y agoI'm curious how much of that is falling out of the top, and how much is the to accelerating ever higher. And you still need to adjust for inherited wealth. That is, of those that fell out, how many got there from inheritance?
- philipov 10y ago1% is a rather arbitrary division point. Even if you assume that there exists a stable cluster of wealth at the top, why would it be bounded around 1%? It is much more likely that the group is either smaller than 1%, or larger. Suppose it's larger, then the group could still be stable, but the churn between people at 1.1% and 0.9% would still cause the arbitrary measurement of top 1% to change a lot. Now suppose the group is smaller than 1%, and the churn is instead at the top of the cluster directly beneath the stable group at the top, but still causes the composition of the top 1% to change constantly.
- taeric 10y agoYes, it is an arbitrary cutoff. If you have a non-arbitrary method to pick the cutoff, propose it. Otherwise, what is the argument?