Using a large sample of proprietary transaction-level institutional trading data, we empirically analyze, for the first time in the literature, the role of institutional investors in corporate spin-offs. In the first part of the paper, we study the imbalance in post-spin-off institutional trading between parent and subsidiary, and analyze this imbalance to test three different hypotheses regarding institutional investors' role in spin-offs: information production, pure play, and risk management. In the second part of the paper, we examine the information production role of institutional investors further by analyzing the predictability of institutional trading around corporate spin-offs for the short-term and the long-term stock returns following spin-offs. In the third part of the paper, we study the pattern and profitability of institutional trading following spin-offs. Our empirical results can be summarized as follows. First, there is significant imbalance in post-spin-off institutional trading between parents and subsidiaries; this imbalance increases corresponding to the difference in the extent of information asymmetry characterizing the two entities, beta risk, and long-term growth prospects. Second, institutional trading in the combined firm two months prior to the spin-off has significant predictive power for the announcement effect of a spin-off. Third, institutional trading in the subsidiary immediately after spin-off completion also has predictive power for its subsequent long-term stock returns; this predictive power is greater when the subsidiary's size constitutes only a smaller fraction of the combined firm's size. Fourth, the predictive power of institutional trading is weaker for the parent firm's long-term returns; further, unlike in the case of the subsidiary, institutional investors start exploiting their private information after the spin-off announcement, but before the spin-off is completed. Finally, institutional investors are able to realize superior profits by trading in the equity of the subsidiary in the first quarter after the spin-off. While our results provide some support for all three hypotheses regarding the role of institutional investors in corporate spin-offs, they provide particularly strong support for the information production hypothesis.