I think he is again, but someone please confirm my intuition.
If you make someone take a break for a quarter of every hour they are working on the job you are roughly reducing their productivity by a quarter. You are shifting the MRP curve to the left. That will give you a new equilibrium where you are going to have lower employment and lower wages.
The same goes for a tax on wages that government imposes on the firm (his initial counter-argument).
Neither of these are at all comparable to the shift along the MC curve that a minimum wage is alleged to accomplish.
Another way of putting it is this: nothing about taxing the firm prevents it from exercising the exact same monopsony power - setting quantity such that MC=MRP and the wage as low as it goes. They just do that with a new MRP curve. Nothing about deliberately making workers less productive (presumably in a way that those workers would like) prevents the firm from exercising the exact same monopsony power - setting quantity such that MC=MRP and the wage as low as it goes. Again, they just do that with a new MRP curve.
Setting a minimum wage does prevent them from exercising monopsony power and as long as it is modest enough it will present them with a wage bargain that would be found mutually beneficial. Should we look at other margins of exploitation? Ya, sure. But that's a different issue.
I feel like I'm taking crazy pills here. Am I? What am I missing about these arguments?