This paper examines two different clearing arrangements for bank liabilities. One was a profit-maximizing private entity, the Suffolk Banking System. It cleared notes for New England banks between 1827 and 1858. The other was a nonprofit collective, the clearinghouses organized in many cities beginning in 1853. The paper examines how well these arrangements prevented bank failures and acted as lenders of last resort. It finds the Suffolk system had fewer failures but acted less like a lender of last resort. It argues that these differences can be explained by the different incentives facing the Suffolk Bank and the members of clearinghouses.

Prior to 1861, several U.S. states established bank liability insurance schemes. One type was an insurance fund. Member banks paid into a state-run fund that paid bank creditors’ losses. A second scheme was a mutual guarantee system. Member banks were legally responsible for the liabilities of any insolvent bank. This paper’s hypothesis is that the moral hazard problem was controlled under a scheme to the degree that member banks had the power and incentive to control or modify others’ risk-taking behavior. Schemes that gave member banks both strong incentives and power were able to control the moral hazard problem better than schemes in which one or both features were weak. Empirical evidence on bank failures and losses on banks’ asset portfolios is consistent with this hypothesis.

The behavior of interest rates under the U.S. National Banking System is puzzling because of the apparent presence of persistent and large unexploited arbitrage opportunities for note issuing banks. Previous attempts to explain interest rate behavior have relied on the cost or the inelasticity of note issue. These attempts are not entirely satisfactory. Here we propose a new rationale to solve the puzzle. Inelastic note issuance arises endogenously because the marginal cost of issuing notes is an increasing function of circulation. We build a spatial separation model where some fraction of agents must move each period. Banknotes can be carried between locations; deposits cannot. Taking the model to the data on national banks, we find it matches the movements in long-term interest rates well. It also predicts movements in deposit rates during panics. However, the model displays more inelasticity of notes issuance than is in the data.

Prior to 1863, state-chartered banks in the United States issued notes–dollar-denominated promises to pay specie to the bearer on demand. Although these notes circulated at par locally, they usually were quoted at a discount outside the local area. These discounts varied by both the location of the bank and the location where the discount was being quoted. Further, these discounts were asymmetric across locations, meaning that the discounts quoted in location A on the notes of banks in location B generally differed from the discounts quoted in location B on the notes of banks in location A. Also, discounts generally increased when banks suspended payments on their notes. In this paper we construct a random matching model to qualitatively match these facts about banknote discounts. To attempt to account for locational differences, the model has agents that come from two distinct locations. Each location also has bankers that can issue notes. Banknotes are accepted in exchange because banks are required to produce when a banknote is presented for redemption and their past actions are public information. Overall, the model delivers predictions consistent with the behavior of discounts.