In theory there may be a point you’re describing, but in SF for instance people just started looting cars instead of mugging (or snatch and grabbing) phones.
That's my whole point. Carrier locking used to be standard, even on iPhones, and phone theft was very common too. Carrier locking wasn't a theft deterrent.
Firmware locking is a great theft deterrent. A stolen iPhone is basically just a few cheaper used parts, and all the expensive ones are useless.
One is extremely effective. One is marginally effective. But that still means both of them are effective just to different degrees.
The only theft it is preventing is other carriers stealing customers from one another.
I am arguing that the reduction in theft due to carrier locks is 0. You are arguing that it is small (marginal).
I am saying that no matter what, the actual cost of the carrier lock to the consumer (as demonstrated by used phone values on eBay), is far higher than whatever marginal benefit you are arguing for.
It is a massive net loss to the consumer, and the only guaranteed beneficiary is the wireless carrier.
My bet is that stealing phones is more like a thresh-hold. Does this crime pay > $x? Then I will do the crime. In this case I suspect that $x is well below the selling price of an unlocked phone. There isn't a gradual dropoff, at least not in the region between $400-$520.
I could also argue that carrier locked phones could have a paradoxical increased effect on theft. If the expected value of stealing a phone is lower, the average thief needs to steal more phones to make the same amount of money.