In the specification for OWL 2, the section on key axioms states:
The semantics of key axioms is specific in that these axioms apply only to individuals explicitly introduced in the ontology by name, and not to unnamed individuals (i.e., the individuals whose existence is implied by existential quantification). This makes key axioms equivalent to a variant of DL-safe rules. Thus, key axioms will typically not affect class-based inferences such as the computation of the subsumption hierarchy, but they will play a role in answering queries about individuals.
It then proceeds to show an example of a set of axioms that would obviously entail a conclusion but that the spec does not recognize, because of that rule.
What are the reasons behind this decision?
The explanation was given in Sec. 3.1.4 in OWL 2: The Next Step for OWL:
Extending DL-based languages such as OWL 2 with keys, however, poses both theoretical and practical problems [23]. Therefore, the Working Group has decided to include a more restricted variant of keys that can be useful in practice as well as relatively easy to implement, commonly known as easy keys.
So they used a restricted form of keys, called easy keys, which has some limitations but doesn'T increase worst-case complexity of reasoning:
Thus, the main drawback of easy keys is that they can only produce consequences about explicit data and are thus relevant mainly for query answering, whereas the general variant of keys [23] can also affect the subsumption hierarchy between classes. The main benefits of easy keys are that adding them to OWL 2 do not increase the worst-case complexity of reasoning, and that implementing them in the existing reasoners is relatively straightforward.