There's a lot of research out there showing that the best preschool programs have enormous benefits to the kids who are in them that, if scaled up, would have enormous benefits to society. Larger-scale programs are hard to study as rigorously (a proper RCT almost by definition has to be small) and not all preschool programs do as well as the best ones, but there's plenty of evidence beyond the Perry and ABC experiments for the benefits of preschool. But preschool advocates, for reasons that aren't entirely clear to me, seem to really enjoy talking about this in terms of the "return on investment" of preschool spending being really high.
This strikes me as an incredibly confusing way to frame the issue. The implication of saying that preschool has a higher return on investment than the stock market is that a state could take its pension fund out of its current investments and use that money to finance a preschool expansion and come out ahead of the game in terms of meeting its pension obligations. That's not true. Or at least there's no particular reason to think it's true.
Preschool has benefits in the same way that taxes have costs. Taxes have costs to taxpayers and taxes have costs to society in terms of deadweight loss and foregone economic growth. Preschool has benefits to kids who attend it and to society in terms of the spillover benefits of more pro-social behavior from those kids down the road. The question is whether spending money on preschool has more benefits than spending money on high school or college or F-35s or bus drivers. Or perhaps the question is whether spending money on preschool has benefits that outweigh the costs of higher taxes. But these aren't financial returns that somehow obviate the normal fiscal issue that to spend a few billion on preschool you need to scrounge up a few billion from somewhere or other.