Societies have incentives to design institutions that allow central bank secrecy. This paper illustrates two of these incentives. First, if society tries to constrain secrecy in one way, central bankers will try to regain lost effectiveness by building up secrecy in other ways. Therefore, we may wind up accepting types of secrecy that appear preventable because reducing them would lead to higher costs. Second, if the social trade-offs between policy objectives change over time, the public may directly prefer greater central bank secrecy so that it will be surprised with expansionary policies when it most desires them.