Copyright cannot apply to facts, like the entity relation "paper X cited paper Y". So this is not a copyright issue.
However, publishers maintain databases of citations to which they sell access. Such databases are protected under sui generis database rights in many countries (e.g. EU, Russia, but not USA) - these are automatic property rights that recognise the effort of compiling a database, similar to but distinct from copyright (in that they don't have to involve any creativity).
Because there is no sui generis database right in the USA, I doubt that's the reason either - perhaps google just don't consider it a priority.
It's extremely unlikely that such database rights are an issue as ISI (Institute for Scientific Information, now part of Thomson Reuters) has been compiling this sort of data since 1960, as well as many other indexing services.
If database rights were an issue then they would have sued each other already and settled things long before Google Scholar.
One thing I've always wondered is why it has "cited by" but no forward citations? Is that a copyright issue?