In fact if we extrapolate that line of reasoning out to the capitalist substrate of our economy we might find some surprising second, third+ order effects.
There's two different questions. Combining the two will give you both answers you'd expect, not one or the other. And there's way too much going on to try and tackle both at the same time.
You could say "defund police marginally increases crime" and "increasing housing greatly reduces crime" but to what extent? And within what timeframes? Defunding the police could provide outcomes immediately, but to provide housing to a point where crime is reduced would most likely be a slow strong growth thing, that will take a generation or two to come about.
I'm not saying it's unimportant to look at both, but looking at them separately will be more realistic, and thankfully money is such an abstract resource that things can be separated like that.
> "We can find some surprising second and third order effects"
I think we don't quite want surprises when it comes to the safety and livelihood of many, and their generations.
It takes centuries to grow a forest and only a day to burn it down.
When for example a lack of housing leads to an increase in crime a well funded police prevents those who could increase housing from being affected by the crime and they will therefor not increase housing. Especially as they get many benefits from poverty like cheap labour, a decrease in competition and larger premiums on attractive real estate.
So yes I would say they are thinking about the second-order effects. At some point I might read some of Thomas Sowell's writings but from what I've seen so far he honestly seems more of a theologist than a scientist to me.