We examine the relation between the agglomeration of firms around big cities and chief executive officer (CEO) compensation. We find a positive relation among the metropolitan size of a firm’s headquarters, the total and equity portion of its CEO’s pay, and the quality of CEO educational attainment. We also find that CEOs gradually increase their human capital in major metropolitan areas and are rewarded for this upon relocation to smaller cities. Taken together, the results suggest that urban agglomeration reflects local network spillovers and faster learning of skilled individuals, for which firms are willing to pay a premium and which are therefore important factors in CEO compensation.

We show that firms led by politically partisan CEOs are associated with a higher level of corporate tax sheltering than firms led by nonpartisan CEOs. Specifically, Republican CEOs are associated with more corporate tax sheltering even when their wealth is not tied with that of shareholders and when corporate governance is weak, suggesting that their tax sheltering decisions could be driven by idiosyncratic factors such as their political ideology. We also show that Democratic CEOs are associated with more corporate tax sheltering only when their stock-based incentives are high, suggesting that their tax sheltering decisions are more likely to be driven by economic incentives. In sum, our results support the political connection hypothesis in general but highlight that the specific factors driving partisan CEOs' tax sheltering behaviors differ. Our results imply that it may cost firms more to motivate Democratic CEOs to engage in more tax sheltering activities because such decisions go against their political beliefs regarding tax policies.

We examine the impact of analyst coverage on corporate tax aggressiveness. To address endogeneity concerns, we perform a difference-in-differences analysis using a setting which causes exogenous decreases in analyst coverage. Our tests identify a negative causal effect of analyst coverage on tax aggressiveness, suggesting that higher analyst coverage constrains corporate tax aggressiveness. Further cross-sectional variation tests find that this constraining effect on tax aggressiveness is more pronounced in firms with lower investor recognition and firms with more opaque information environments. Our results are consistent with the notion that higher analyst coverage increases the visibility of aggressive tax planning behavior as well as heightens analysts’ demand for more transparent information, which in turn reduces tax aggressiveness.