> for every person who leaves the middle class to become poor, more than one person leaves to become upper middle class.

I'm not sure this is true, and the only way I can see it possibly be is because "middle class" is a very fuzzy category that's defined relatively to others, so it can grow in absolute terms by e.g. lowering that relative standard. Income inequality has steadily grown, not shrunk. Concentration of capital at the top, and concentration of income from that capital, has also grown.

> not how to make the well off less valuable through arbitrary productivity restrictions.

You're looking at it the wrong way. It's not about restrictions on productivity. It is about redistribution of capital, and prevention of it getting concentrated too much in one hands at the point where labor is going to lose access to more and more of it to feed themselves via wages. I don't even thing that capping ownership of capital is the right way to do so. Taxing the hell out of it would be a better idea.

And education and training is not going to do anything to the fundamental problem of 100 people competing for 50 jobs.