^The important part is to have consistent methodology.
Mr. Hulbert's data seems to include direct and indirect ownership.
One can use data from the Federal Reserve Financial Accounts but you have to decide what to include. Some items are straightforward ie direct holdings of corporate and listed equities. You have to decide if indirect ownership through pension assets needs to be accounted for and it can tricky with 'balanced' funds which contain a variable percentage of equity exposure.
Some of these findings come from well done consumer surveys but this would tend to be less precise (although possibly accurate).
On top of that, you may want to adjust for other demographic factors. The US is getting older: median age in 1990:32.9, 2000:35.3 and 2019:38.4. Also, you may want to remember (or ignore) that stock ownership (especially since 1999) is becoming more concentrated in the top 10% and the top 10% of that 10%. Since the GFC, stock ownership (along age and income profile) has significantly decreased for all groups unless you're 65+ or make more than 100K a year. Interestingly(?), in 1999, the middle class held about 25% more wealth than the top 1% and now the top 1% holds 50% more wealth than the middle class (the numbers reported here also involve some reasonable methodological choices about definitions etc). i don't really care about this last part now but i wonder if this has future implications when we-the people eventually start to claim their share of the we-1%.
