Not needing to store hash values greater than V(kth) means we can automatically reject any incoming hash values greater than or equal to V(kth) up front.
Additionally, we can reject any incoming hash values that are duplicates of any of the hash values we already have in our KMV list.