Abstract: Startup founders, who generally must cede control to obtain VC financing, are widely believed to regain control in the event of IPO, à la Facebook’s Mark Zuckerberg. Indeed, the premise that founders expect to reacquire control if there is an IPO underlies the leading finance theory for why venture capital cannot thrive without a robust stock market: the existence of an IPO market enables VCs to give founders a “call option on control” exercisable if the firm is successful. But little is known about how frequently founders regain control via IPO. Using a sample of over 18,000 VC-backed firms that received their initial round of VC financing during 1990-2012, we show that founders generally do not reacquire control if there is an IPO. In almost 60% of firms that do go public, the founder is no longer CEO at IPO. In firms with a founder-CEO right after IPO, founders generally lack substantial voting power; 50% are no longer CEO of the firm within three years. Zuckerberg is not the norm. We also show that the ex ante likelihood of any given founder reacquiring control via IPO is extremely low, suggesting that the expected value of any call option on control is trivial. As of initial VC financing, the likelihood that a founder takes her firm public and retains the CEO position and voting control for three years is about 0.4%. Our results shed light on how control evolves in U.S. startups, and cast doubt on the plausibility of the “call option on control” theory linking stock and VC markets.