My point is that older people with longer memories can still note that today's crime rate is by no means low, even if younger people (say, people my age or younger) have lived in a United States with declining crime rates for more than a decade. Trends are important, but what base they start from is important as well.
As I read your post, I was thinking that increased connectivity between individuals (I'm now in contact with many people I'd have lost contact with ages ago, thanks to FaceBook etc.), together with the widespread mistake that "anecdote is the singular of data", combine to create a perception that more bad stuff is happening.
That is, I'm in contact with many more people now than I could have been a few decades ago: everyone I associate with now, plus many people I've reconnected with via social media. I daresay that connects me with more people than, say, my grandfather was connected with. And if my larger sample hits one crime, then our mental inability to scale creates a false impression that the odd event is happening all around.
