The day before yesterday, I had a new student scheduled to have a trial lesson at 4:00 pm. When the doorbell rang at 3:00 pm, I thought that the fellow had gotten confused and showed up an hour early. I answered the door and a tall, moon-faced policeman stood there, looking a bit like a retired sumo wrestler who'd managed to lose all his fighting weight, with a large ring binder in hand. This was a routine I'd been through at least twice before, possibly more, so I wasn't all that concerned. In fact, it was this very same policeman who had been at our door in the past.
In Tokyo, and perhaps in other areas as well, the local police go around from door to door in the neighborhood they work in with a binder collecting information. They write down the number of people who reside at each address, their ages, their jobs, and the type of place they work at. In our case, the information hadn't changed so it wasn't very complicated to communicate our situation to him. This evening, the same policeman was parked outside our apartment building around 7:00 pm and was trying to pick up information on the residents of our building who hadn't been home yesterday afternoon (those on the second floor). I passed him on my way home from a store and he thanked me for cooperating with him yesterday and said "good evening" (all in Japanese, of course). After finishing with our building, he trotted off down the street to other buildings nearby.
I'm not sure how often this type of information is collected or updated but I'm guessing it's once a year. On the surface, this all seems like a pretty good idea. On the one hand, the police get some idea of who does and does not belong in an area based on their overview of the residents. If they are patrolling on their bicycles or on foot, they can note suspicious activity if they see someone who doesn't seem to belong based on their data. On the other hand, I don't know if this information is actually put to any use or if it's just one more of those things the Japanese do meticulously as a matter of course. It's possible it's all just dotting the "i's" and crossing the "t's". If I had to guess though, I'd wager the information is coordinated with witness testimony in the event that a crime is committed.
Unfortunately, I very much doubt this does much good in the case of most common local crimes. In one of my lessons, I discuss "tansu yokin" with students. "Tansu yokin" is "mattress banking" in Japan. That is, it is akin to the concept in the west of savings one's money under the mattress instead of putting it in a bank though, in this case, it's usually a stash of cash, jewelry, a personal seal ("hanko") used for banking transactions or other valuable items. During this particular lesson, one of my newer students told me that her house was burgled and her family's stash was stolen. When they reported the incident to the police, they were told that three houses had been burgled in the same area on the same night. It seems that the thief (or thieves) cased the area and were able to rob several houses effectively while the residents of those houses were at work on their regular schedules.
Despite the fact that the robbers clearly were around the same area for the duration of committing three robberies, there wasn't enough evidence to do anything about it. In fact, the police told her that it was highly unlikely that they would find the criminals or retrieve the loot. So, even knowing the neighborhood (bear in mind this student lives relatively close to where I do), wouldn't appear to help, particularly if no one notes the appearance of unfamiliar faces while crimes are occurring.
Nonetheless, I do feel it's rather a good thing to have the police become familiar with the people who live in the neighborhoods they are stationed in. If nothing else, I'm guessing the policeman who wished me a good night and came to our door to collect data won't be stopping me to see if I've stolen my bicycle. ;-)