I am wondering if there has ever been a study, academic or commissioned by the USPS, on what percentage of First-Class Mail goes undelivered for some length of time. I’ve searched high and low, but no luck. You’d think this is something the Post Office would want to know, but maybe they don’t want us to know. I now turn to you, Straight Dopers, because I know that where Google fails, you triumph.
According to this article “over 100 million pieces of mail” annually are undeliverable in the U.S.
Right, but I’m looking for info on what fraction of properly addressed mail is simply lost - mail recovery centers are generally for items that don’t have proper addresses. It seems like an easy enough thing to do; I’m surprised that I’m unable to find any proper study on it. Surely the USPS does some sort of quality checking.
Well, for that it seems you’d have to know exactly how many pieces of mail went into the system, and seeing that there isn’t any easy way to tag and track all types of mail from the moment it leaves someone’s hands until the moment it arrives, that would be a tall order.
Yes, you could track how many pieces of registered or delivery-confirmation mail arrived, for instance, but you would still miss out on what pieces of registered mail went missing before they ever got scanned into the system, plus special mail like that of necessity draws attention and requires special handling, and should have a higher success rate at being delivered.
I’m sure someone out there has done studies involving test mailings or something similar, and they’re just buried somewhere amidst dull statistical analyses and incomprehensible jargon.
Yes, I was talking more about test mailings . . . obviously tracking every piece of mail would be impossible.
You could do a study where you have planned mailings between a sufficiently large set of endpoints all around the country and/or world. Then send the mailings and track what percentage never arrive after a reasonable time plus a meaningfully large buffer time for burps in the system.
Of course, one obvious problem with that is that if the postal service figures out what you are doing, then they may be tempted to pay extra special care to mail addressed from or to a known “study” address above and beyond the care given to normal mail.
According to this paper (warning, pdf) “less than 0.5%” of mail is reported lost in the U.S. and U.K.
I could have sworn I saw a recent thread that linked to a paper about an experiment where the experimenters tried to see what portion of undeliverable mail (sent to addresses in a large number of different countries) with a valid return address would actually be returned to the sender.
That works out to (on average) over 5000 pieces every single day in every single state (or about 5 pieces per office per day (but that doesn’t quite work since things get transferred from office to office)).
The USPS does surveys of delivery and hire firms to test by sending mail to consumers, who report what mail arrives. IIRC, when I was taking part, the number of lost mailpieces was very small – certainly less than 1%.