Personally, I think there's two factors at work here.
1: Motivation to buy new. If you think about it, who are the people most likely to rush out and grab a band's latest release as soon as it comes out? The people who are already fans of that particular group, of course! And so it's only natural that their reviews will tend towards the higher ratings - well, unless a band tries out a different approach with a new album which alienates a large proportion of their existing fanbase, which does happen, but I don't think it happens so often as to throw the trend off.
2: Motivation to review. It's always easier to push yourself to write glowing praise of an album you really, really love - the urge to share the joy with people is a powerful one. (Likewise, if you seriously dislike an album that could also be a good motivator to warn the world about it...) It's a bit harder to motivate yourself to review an album you're only lukewarm about or which you consider to be OK but not especially praiseworthy or horrible. So in general people are a bit more likely to review albums they want to rave or curse about than albums they don't particularly care about.
When you take those two factors and put them together, the people reviewing albums when they are newly released are disproportionately likely to be folk who a) like the band in question and were looking forward to the release and b) formed a fairly immediate opinion about the album rather than being unsure and wanting to give it a few more listens before reviewing. So it's only natural that new releases get good reviews and then get harsher criticism later on.
Personally, I think my 5-star ratings are a bit high at the moment, but they're skewed a bit because at the moment I'm doing a project of going through my music collection in chronological order and reviewing *everything* - which means that the albums I'm reviewing now will tend to be the 1960s-early 1970s albums which have well and truly stood the test of time. I anticipate my proportion of 5-stars will go down over time as I get to more recent albums which perhaps are a bit more imitative and a bit less innovative.
What I think might help is a rule that an album can't appear on the top 100 lists until at least 6 months after its release date; I think that's a fair amount of time to let the initial wave of enthusiasm die down and get some input from cooler heads.