> Government spending on production is usually mis-allocation, because if it was productive a private company would be already doing it.

This is learning the wrong lesson from broken windows / TANSTAAFLing yourself into thinking "all spending has the same ROI and it's never positive."

Private companies will only do things that accrue economic gains to themselves instead of returning those economic gains across the population as a whole.

I didn't say "all" I said "usually". Making up things and then arguing against them is the strawman logical fallacy.

I interpreted your "This extra spending is paid for with inflation." perhaps more generally than you meant it, then. "Any extra spending would've led to inflation" vs "that specific extra spending led to inflation" - I read you as more the former, not the latter.

Re: "usually" I'm not sure if I disagree with it because I don't know what you mean. "Usually" in practice or "usually" in possibility? By dollar share or by project share or other? And how directly are we measuring ROI, especially in terms of infrastructure or technology development from large projects?

I believe we could certainly get to a point where most government spending was net beneficial without needing to shrink the services provided by the government.

EDIT: and if we're picking on logical errors instead of getting into more details: my claim was that other forms of spending could avoid inflationary problems. I don't see your response that this would usually be mis-allocation or unproductive saying anything directly about my claim.

If the government prints money and spends it on things that are less productive than the free market would, it results in inflation.