The dissertation, which consists of three chapters, is devoted to exploring theoretical asset pricing in over-the-counter markets. In Chapter 1, I study an economy where investors can trade a long-lived asset in both exchange and OTC market. Exchange means high immediacy and high cost while OTC market corresponds to low immediacy and low cost. Investors with urgent trading needs enter the exchange while investors with medium valuations enter the OTC market. As search friction decreases, more investors enter the OTC market, the bid-ask spread narrows and the trading volume in the OTC market increases. This sheds some light on the historical pattern why most trading in corporate and municipal bonds on the NYSE migrated to OTC markets after WWII with the development of communication technology. In Chapter 2 (co-authored with Hongjun Yan and Hin Wei), we analyse a search model where an intermediary sector emerges endogenously and trades are conducted through intermediation chains. We show that the chain length and the price dispersion among inter-dealer trades are decreasing in search cost, search speed and market size, but increasing in investors’ trading needs. Using data from the U.S. corporate bond market, we find evidence broadly consistent with these predictions. Moreover, as the search speed goes to infinity, our searchmarket equilibrium does not always converge to the centralized-market equilibrium. In particular, the trading volume explodes when the search cost approaches zero. In Chapter 3 (co-authored with Hongjun Yan), we analyse a search model where two assets with different level of liquidity and safety are traded. We find that the marginal investor’s preference for safety and liquidity is not enough to determine the premium in equilibrium, but the whole distribution of investors’ valuations play an important role. We specify the condition under which an increase in the supply of the liquid asset may increase or decrease the liquidity premium. The paper also endogenizes the investment in the search technology and conducts welfare analysis. We find that investors may over- or underinvest in the search technology relative to a central planner.