This article uses a general equilibrium framework to explore the origins and limitations of financial intermediaries. In the model, investors have a generic lending technology that they can improve at a cost. Those who upgrade become intermediaries to exploit their advantage. However, conflicts with depositors will limit the banks' market presence, and they will only lend to moderately endowed firms while bondholders will finance cash-rich corporations. The article also analyzes the extent to which investors adopt the superior lending technique, the nature of bank competition, and how corporate and bank conditions affect interest rates and investment.

This article studies how equity ownership and corporate control were separated in the United States. Initially, railroads and industrial firms were tightly controlled by a few shareholders; this situation was altered in the 1890s by massive mergers and reorganizations, which allowed private banks to control railroads and industrial firms. Between 1912 and 1939, bank control faded away as a result of a political reaction against financial institutions. Using stock-market data from 1914, the author shows that the eviction of banks from corporate boards depressed firm values by about 7 percent and that part of this value came from cartelization.

This article investigates which companies finance themselves through intermediaries and which borrow directly from arm's length investors. Our empirical results show that large companies with abundant cash and collateral tap credit markets directly; these markets cater to safe and profitable industries, and are most active when riskless rates or intermediary earnings are low. We show that determinants of lender selection sharpen during investment downturns and that there are substantial asymmetries in the way firms enter and exit capital markets. These results support a theoretical framework where intermediaries have better reorganizational skills but a higher opportunity cost of capital than bondholders.