This paper examines seven instances in which the market value of a parent company was less than the market value of its holdings of a publicly traded subsidiary. Efforts are made to explain this “parent company puzzle” in terms of taxes, agency costs, liquidity effects and noise trader risk. None of them work. The only explanation consistent with the evidence is a mispricing of the subsidiary shares associated with a downward sloping demand curve. As further evidence in support of this view, five corporate control transactions, all designed to exploit the apparent mispricing, were initiated while this research was in progress.