Yes, if we measure income inequality with a one year snapshot, we'll find greater inequality than if we measure it using total lifetime earnings. That's pretty much basic statistics.
I don't see that Talib gives a source for this data. Here's an article that quotes the same number and at least gives a researcher's name.
https://money.cnn.com/2016/01/07/new...p-1/index.html "Thomas Hirschl, a sociology professor at Cornell University"
But, I couldn't find the original research.
I'll guess that if we do, we'll find that a good chunk of the variability is driven by a one time capital gain.
If Hirshl used IRS data, it would be possible to do multiyear sums. Looking at his Cornell web page, he might have done some of that, but I haven't found it.