A Price-Differentiation Model of the Interbank Market and Its Empirical Application (Job Market Paper) [A Complete Appendix Will Be Posted Soon]
Abstract: We study how banks lend or borrow liquidity (central bank money) in the interbank market and what we can learn about the macro-economy from the interbank market. From a unique database of interbank loan transactions in Mexico, we observe that interest rates vary across different lender-borrower pairs. We find that this variation is driven by the variation across different banks in their cost from handling an excess or a deficit of liquidity (`liquidity cost'). We characterize the shape of the interest rate curve as a function of loan size and find that a small bank that trades with a large bank tends to get better interest rates for larger loans. The model is consistent with the following empirical observations: (i) the interest rates on the loans between two large banks are very close to the central bank target rate; (ii) small banks experience a disadvantage against large banks: Small banks lending to large banks receive interest rates lower than the target rate most of the time; and small banks borrowing from large banks pay interest rates higher than the target rate; and (iii) a small bank trading with a large bank obtains more favorable interest rates for larger loans. Finally, as an application of the model, we discuss how banking environment changes during a financial crisis. In particular, we estimate the shape of the liquidity cost function and use that information to measure the shift in the liquidity cost that banks faced during the 2008 financial crisis. We find that the increased disadvantage that small banks experienced in the interbank market during the crisis can largely be explained by a shift in the liquidity cost, rather than by changes in loan supply and demand.
Research in Progress
Payment System Design in a Model of Money (With Robert Townsend)
Abstract: We study a model of money as a medium to facilitate exchange in the context of modern banking systems. Different general equilibrium models of money as a device to facilitate exchange of goods have been proposed where the need for money arises for different reasons. A few well-known ones include geographical separation and seasonal variation in economic activity, for example. While some of the motivations for the use of money in the earlier literature such as geographical separation may not be applicable to a modern banking system due to a drastic decrease in communication costs, we propose that there still exist forces that can motivate the use of money in a modern banking system such the settlement of multilateral trades and unverifiable future trading plans. In particular, we relate the outcome of our study to the design of the payment infrastructure between financial institutions. The real use of money as a medium of exchange is not simply defined by the property of money itself, but by the structure of the payment system within which the money is actually used. The modern banking system is interesting in this regard because there exist important devices such as an intraday overdraft facility to enhance the use of money. We evaluate the welfare properties of existing payment infrastructures under our framework