The determinants of large compensation for executive and non-executive board members are still debated. Bertrand and Mullainathan QJE 2001, using Excucomp data, find little role for firms' size and attribute the large growth to luck. Subsequently Gabaix and Landier QJE 2008 developed an equilibrium model of CEO pay with positive assortative matching and two-sided heterogeneity. CEOs have different talents and are matched to firms in a competitive assignment model. In equilibrium the CEO's pay depends on both the size of the CEO firm and the aggregate firm size. They test the implication of their model, also using Execucomp data. They find that firms' size matters, both the one of the CEO and the aggregate size of firms. Still size does not seem to explain all of the compensation. We conjecture that an additional component, that has received little or no attention so far, lies in the role of network connections. Those connections should play an even larger role in the contemporary social media-based societies. We will focus on two countries, namely US and Italy,. The choice of those two countries is not coincidental. In fact in both there have been recent changes in inter-locking regulations, namely the number of boards an executive or non-executive member can be in. Those changes in regulation represent a good experiment for testing the role of changes in ties to companies and other directors. Our empirical analysis will make use of BoardEx data and in hand collected information on CEO compensation, the latter being based on SEC documents for the US and Bank of Italy documents for Italy. Beyond the empirical analysis, we will write a model to rationalize our evidence. Specifically, we extend a model with two-sided heterogeneity and positive assortative matching (along the lines of Shimer and Smith Econometrica 2001) by allowing for CEOs bargaining power to depend upon the extent of network interconnection. Within the wage norm given by standard Nash bargaining such modification naturally leads to larger shares of the surplus assigned to CEOs. Given the heterogenous distribution of CEO talents and firms' characteristics, our model will also generate implications for wage inequality. We will at last match the model implications in terms of wage inequality and CEO premia by matching the data at hand.