QR codes support multiple devices at the same time, and have much larger range for reading (limited by the physical display size). Unique ID's can be built into the app reading the QR code for things like payment transactions. If I control the app reading the QR code (like WeChat), I can easily add state and add some "ping-back" behaviour to ID who/what/where the scan happened.

I think you misunderstood. I never talked about ping back or anything like it. For that, both are exactly the same. I was purely talking about how it scales when you need to identify each item individually, for instance library books. But it wasn't exactly clearly communicated, my bad.

For inventory management, it's preferable that each item has its own unique code. So either you need to print a different QR code for every item, or you put on an NFC chip.

It is true that QR codes are supported on more devices, but that is because cameras and screens are ubiquitous with smart phones. An NFC reader is cheaper.

I'm not convinced that it'd be any harder or more expensive to print a series of unique QR codes onto objects than to use NFC chips.

The heavy use of (granted, not strictly QR) 2D barcodes in physical mail/shipping suggests it's both useful and cost effective.

Indeed, but they are already printing a label for every parcel, and they used barcodes long before NFC tags was a thing. They already have hardware and tooling in place for it, so unless NFC will save them a significant amount of money it makes no sense to switch.

Think of a stockroom. When you get new inventory, you put on an NFC tag on every item, and then you have to register each which is just one sweep. With QR/barcodes you either have to register and then print and put on the items, or you have to scan each individually after.

Both NFC and QR have strengths and weaknesses. I'm certain that we will see both utilized more and more.