I show that the presence of a lead independent director on the corporate board is positively associated with investment efficiency. The result is more pronounced for firms with weaker corporate governance standards, less transparent financial disclosure, and greater financial constraints. The lead director presence is negatively associated with overinvestment (underinvestment) for firms with large cash balances and low leverage (high cash flow volatility). Moreover, the lead director investment-related committee membership as well as CEO power matter in this setting. The lead director board role is also positively associated with future firm performance.

We examine the impact of CEO social connections on the value of real options. Consistent with the benefits of information transmission, reputation, and trust embedded in social connections, we find that CEO social connections act as real option facilitator through improved financial flexibility, alleviating financial constraints. The effect of CEO social connections is stronger for firms with high growth opportunities and when the CEO has a longer-term career horizon and greater ability. CEO social connections have an increasingly stronger impact on the value of real options. Overall, social connections influence the value of real options, a novel finding in the literature.

We provide the first direct empirical evidence of the effect of CEO social capital on aggregate corporate risk-taking. Our theory predicts that CEOs with high social capital display higher levels of risk-seeking behavior. Consistent with this prediction, we find a positive association between CEO social capital and aggregate corporate risk-taking. Examining the channel, we show that social ties cause corporate policy actions, and these actions lead to greater volatilities in stock returns and earnings. In addition, we uncover a number of factors that significantly moderate the effects of social capital on risk-taking. We also show that this increase in risk-taking is value-enhancing to the firm. Our results are robust to alternative proxies for risk-taking, alternative model specifications, and tests for endogeneity.