It is absolutely an informal fallacy[1]. The onus is on the person making the claim of similarity to demonstrate why that similarity is valid, so on its own it lacks any argumentative weight.

Something being an informal fallacy doesn't mean that it can't be a component of a good argument (ie. correlation not implying causation doesn't mean there are no cases where something can be demonstrated to be causitive), it means that it is not an argument in and of itself.

[1] http://en.wikipedia.org/wiki/Informal_fallacy