This dissertation contains three essays exploring the impact of network structures and information frictions on aggregate conditions. In Chapter 1, we consider a banking network formation model where banks can insure against default risks and learn about the underlying state of the economy through their network. Our findings show that banks are less connected compared to the socially optimal network due to positive externalities. We suggest that adjusting the network formation cost can enhance the efficiency of the network on equilibrium. Along with networks' externalities, in Chapter 2 with Wentong Chen, we provide empirical evidence indicating that US firms are over-hedging against foreign risks, overlooking potential indirect hedging benefits from upstream or downstream sectors. To align with the observations, we incorporate hedging decisions into the standard New Keynesian framework to explain firms' hedging strategies. In Chapter 3, we study the aggregate consequence of network-dependent information on the effectiveness of monetary policy. Based on derived firms' nominal pricing, numerical experiments shows that a more connected network exhibits a more gradual response to monetary shocks compared to a less connected network. Additionally, with productivity shocks, the effects of information frictions become more persistent over time as firms cannot identify whether the observed fluctuations originate from monetary or productivity shocks.