Beta anomaly and real anomaly

Description

Understanding corporate investment decisions is crucial for academics and practitioners. Intuitively, a high cost of capital deters investments because it heavily discounts future cash flows generated by those investments. The cost of capital, which serves as a discount rate, increases with the riskiness of investment, because investors do not like risk. Therefore, the higher the risk, the greater the cost of capital, and the lower the investment. However, in this proposal, we document a novel, robust empirical result: firms with high systematic risk effectively invest more than those with low risk. This evidence is contrary to the intuition discussed above. Moreover, this result challenges the so-called “real option” effect, which refers to the idea that firms wait to invest when uncertainty is high. To understand this puzzling result, we utilize a well-known empirical observation in the asset pricing literature, the beta anomaly, to understand the puzzling positive relation between systematic risk and abnormal corporate investments. The beta anomaly refers to an empirical finding that firms with high systematic risk (or “beta”) receive a low risk-adjusted return (or “alpha”), on average. Therefore, we conjecture that high-risk firms might take advantage of the low cost of capital and invest more. To verify the conjecture formally, we build a dynamic model to help us understand how the cost of capital determines investment. Specifically, in our model, the cost of capital includes a risk-based component, which increases the cost of capital, and a mispricing component, which is novel to the literature. With the new ingredients, we aim to achieve two objectives. First, we assess whether the mispricing component in the model can dominate the risk-based component to generate the low risk-adjusted cost of capital for high-beta firms. Second, the managers of high-beta firms make investment decisions instantaneously, when observing the low cost of capital. Therefore, our model is expected to replicate the empirical regularity: the positive relation between high systematic risk and corporate investment. In short, in this proposed study, we plan to provide a unified model to understand the negative beta-alpha relation (the beta anomaly) and connect it with the positive beta-investment relation (the real anomaly).