I would challenge the idea that all older pets are rarer by saying that we don't know the ratios pets are set at upon their release. Take this litter for example:
archive/2016/March/Dog
We don't know what ratio this pet was released at. The only way we could accurately determine if this pet is rarer would be a rarity change; whichever pet changed to VR first would be more rare. So, a 2016 rare could absolutely be rarer than a 2010 rare. Generally we assume that older pets are rarer because less users played CS and less pets were adopted, but I'm sure that there are exceptions to this rule. When pets are re-released, they are done so in a way that maintains their rarity (the exception to this being a pet that is bordering on a rarity change reaching the "tipping point" due to a re-release, like we've seen with many July '08 litters).
I think a good rule of thumb is that yes, pets released in 2008-2010 are rarer than new pets. But when rare and VR pets are introduced in monthly releases, then we really don't know where they lie.
And what about the August PPS from August 2012? We don't know where it lies on the scale either. It could be worth more than an older VR pet, or it could not be. We just don't have that information.
Rarity tags are broad to discourage users from being too specific about their trades. We can only make accurate claims when we have rarity changes to back them up.
Also, ZEL mentioning that the Moonswirl bunny was VR before the CEG blew my mind. I think it just goes to show that rarity changes are the most reliable way to determine values within any rarity label. I am very surprised!