I mean, they already do exactly this — this is what Amazon Lockers are. It's just only seemingly worth it to Amazon to deploy them to commercial customers, e.g. at post offices, in front of Whole Foods locations, in some very large apartment building complexes, etc.
(My own guess as to why the economics don't work out for individual residences, is that a hypothetical smaller locker — one small enough to fit on a porch — would also inherently be lightweight enough for thieves to just cart away wholesale.)
Every country has these to some degree; I imagine they're most popular in places that 1. have colder climates, but 2. where people don't tend to drive (like Poland?) The US has some, but the suburban long-distance-commute car culture + generally not-too-bad climate, means that people in the US generally expect to pick up packages from further away, and so implementation of these in the US has lagged behind other countries.
However, my comment, and the one it was replying to, are talking about something else — a hypothetical concept of small lockers that serve single homes, given to the homeowner, to be located near the home's mailbox/mailslot. (Basically, logistics-provider-provided versions of these things that you can technically buy online — but where I've never seen anyone with one: https://www.amazon.ca/WeHere-Package-Delivery-Anti-Theft-Pas...).
And the thing about these is... they really aren't a good idea. They're not too big and heavy to just steal. Anyone who can walk up to your porch with a moving dolly can walk away with it.
I agree that per-household lockers are... tricky at best. But then, if we're talking homes, and thus presumably lawns in front of them, I wonder what are the difficulties of selling a multi-slot locker that would be bolted down to the ground (or perhaps a bunch of concrete filling a hole in the ground), and thus as easy to steal as a thick fence post or an ATM? Is this too expensive for homeowners?
I suspect that some of the the same reasons that lockers aren't economically feasible in most of the US is the same reason that the theft isn't a problem: low density. If you're a porch pirate, you need to expect that the value of your stolen goods covers at least your gas and time driving around stealing stuff, plus some risk premium for doing the crime. If the average value of a package is below this amount, the crime doesn't pay. There will still be instances where people haven't done this math, or crimes of opportunity, or just dense stretches where it does make sense depending on the price of gas, but it isn't a nationwide problem.
I get hundreds of packages per year (not an exaggeration) and as far as I know, exactly zero have ever been stolen. Missing packages are invariably delivered to someplace else that must have had a better vibe for the driver that day. (I’ll get pics of proof of delivery with a package that is clearly not at my house.)
In that environment, what problem do I have that could be solved by this, and how much effort (and aesthetics) am I willing to spend to solve it?
Now, if my house shared a wall with two other houses and people walked by my front door all the time, maybe I’d have a theft problem due to greater opportunities for it to happen.
Places with lawns probably don’t have nearly as much package theft just due to less foot traffic.
When this happens you have to go to a post office to get your mail.
Some older houses have passthroughs built into the walls for deliveries of milk or coal or ice. I’m surprised this feature hasn’t been resurrected for package deliveries.
