Banking Market Structure, Risk, and the Pattern of Local Interest Rates in the United States, 1893-1911

Abstract

IN the postbellum United States a number of mechanisms existed which promoted the interregional transfer of short-term capital. For example, the correspondent banking system linked country banks with city banks in a pervasive network, as indicated by the fact that in 1913 10 selected New York banks alone held over 15,000 accounts of out-of-state banks (Beckhart and Smith, 1932, p. 156). Slightly later, in 1925, 600 out of 655 Georgia banks maintained accounts in New York; while as far away as CaliforRia, 515 out of 644 banks did (Watkins, 1929, pp. 408-411). These city correspondents purchased commercial paper for country banks, giving even remote banks access to national funds markets, and also transferred funds to them through interbank lending and rediscounting. Although interbank borrowing amounted to only about 112 % of total national bank loans and discounts in 1892-1897, it was an important source of funds for a region like the South, where it constituted at least around 10% of total loans and discounts (Breckenridge, 1898, p. 137). Other methods of interregional funds transfer existed outside of the correspondent banking system, such as direct interregional lending. In 1915 almost 30% of the loans of eastern reserve city banks were made interregionally; almost one-half of all loans in the South made by reserve city banks in 1915 were made by banks outside the South (U.S. Comptroller of the Currency, 1915, pp. 18-19). Interregional holdings of bank stock also represented important sources of funds in some areas; for example, over the late nineteenth century between 25% and 40% of total national bank shares in the Great Plains and western states were held by investors outside the state. Finally, the commercial paper market facilitated the transfer of funds from lenders to borrowers in different regions. Although there were transfers of funds among regions,1 nevertheless substantial interregional interest rate differentials in realized as well as in quoted rates existed.2 These differences were taken as evidence of an imperfect national short-term capital market and of the existence of local monopoly power in southern, midwestern, and western states. Explanations of the movement toward a national capital market over this period have all assumed the existence of barriers to interregional capital mobility. To Lance Davis the westward spread of the commercial paper market facilitated interregional transfers of funds; Richard Sylla, on the other hand, has emphasized the role of high minimum capital requirements and other legal barriers to entry as supports of local bank monopoly power before the more liberalized requirements of the Gold Standard Act of 1900 took effect.3 Risk, however, has to be taken into account explicitly; even in a perfect capital market local interest rates may diverge if differences in risk across regions exist. Were the existing institutions for the interregional transfer of short-term capital in the postbellum period adequate for the operation of a well-functioning national money market? In other words, did a perfect national capital market exist at that time? In order to separate Received for publication June 30, 1975. Revision accepted for publication February 3, 1976. * I am greatly indebted to Peter Temin and Richard West for their many helpful suggestions and comments and to R. M. Hartwell for his diction. This article originally constituted part of a much more lengthy paper presented at the Cliometrics Conference, Madison, Wisconsin, April 1975, and at the economic history workshop of the University of Chicago. Comments from participants at these sessions are also gratefully acknowledged. Responsibility for any remaining errors lies with the author. 1 For a rough estimate of the magnitudes of long-term and short-term interregional capital flows over the 1900-1910 decade, see James (1974), pp. 200-212. 2 For a contemporary discussion of this phenomenon, see Breckenridge (1898). 3 See especially Sylla (1969) and Davis (1965).