Anything metric that becomes gamed ceases to be useful. If we all know the game is to optimize for number of citations, then the jig is up. It will eventually cease to be a useful system and fail its ultimate goal. Institutions can handle some amount of it, but left unchecked, the institution will fail when quality falls enough to prevent any further value creation. Certain journals or periodicals will lose their notoriety, and with it, bring down the credibility of the entire system in which they exist.