> ...this would at least be breathing room.
It might just be the analogy talking; but that sounds like the plan to fix a problem that does not actually fix the problem. It may be a bad plan.
> $462b could be put into an education or infrastructure trust, and pay dividends towards our schools, public housing, roads, and transit networks.
If you think about it, you may discover you are describing what already happens. Your proposed solution is the fund managers are ... y'know, some bureaucrat ... and some % of the fund goes to government coffers. The current situation is the 1% are fund managers and the withdrawal rate is linked to their taxes paid. At a very high level all that proposal does is fiddle around the edges with what the effective rate of withdrawal and what gets invested in and it is entirely possible the total take will drop. The wealthy currently invest in housing, education & transit very visibly.
It is very unlikely that a government fund would survive without attracting a serious wave of leeches; the incentives to raid the fund and waste the money would be huge. The process is fraught. It would be far safer and probably more effective to look to incentives rather than subject investment to the erratic outcomes of government voting.