Printer Friendly


One of the hallmarks of Eugene White's scholarship is his knack for using detailed historical examples to raise large, thought-provoking questions. "Were Banks Special Intermediaries in Late Nineteenth Century America?" is no exception. White combines case studies of two small financial institutions, the Bank of A. Levy and the Emigrant Savings Bank, with other information on nineteenth-century banking theory and practice to highlight banks' "specialness" during that period - that is, their unique ability to serve as delegated monitors for savers. He then argues that a fall in the cost of information eroded this specialness in the twentieth century. The result, according to White, was a steep decline in banks' share of the assets of financial intermediaries.

My goal in this comment is less to criticize White's argument than to amplify and recharacterize it - to make stronger his case that, by the early twentieth century, banks had lost many of the attributes that allowed them to perform the role of delegated monitors more effectively than other kinds of financial institutions. I will begin by countering White's case studies with the example of a late nineteenth-century institution, the Suffolk Bank of Boston, that had largely abrogated its position as delegated monitor. I will then use what is known about the evolution of managerial practice within banks in the late nineteenth and early twentieth centuries to argue that the Suffolk example is more representative of historical trends than the cases White describes. Banks' earlier informational advantage had derived from the imbeddedness of their officers in the communities within which they did most of their lending. As credit markets lost their local character, these advantages disappeared, and banks increasingly had to rely on the same general information sources as other financial institutions.


White uses the examples of the Bank of A. Levy and the Emigrant Savings Bank to highlight the information services that, he argues, were at the heart of banks' specialness in the late nineteenth century. The former institution was a small rural commercial bank in Ventura County, California. Its president, Achille Levy, knew borrowers personally and traveled around the county on horseback in order to monitor their activities. The second, a mutual savings bank in New York City, was run by prominent members of the Irish immigrant community who presumably were personally acquainted with the mortgagees to whom they lent the bulk of the bank's funds. White makes no pretense that these cases are representative, but he does imply that they capture in important ways the kind of information-gathering facilities that made banks special.

The problem is that it is easy to offer counterexamples. One bank that appears to have behaved very differently was the Suffolk National Bank of Boston. During the first half of the nineteenth century, as Rolnick, Smith, and Weber describe in their contribution to this volume, Suffolk was the most important bank in Boston and exerted what was in effect regulatory authority over the notes issued by all the banks in the New England region. This regulatory role ended before the Civil War, however, and by the late nineteenth century, the Suffolk was just one of a considerable number of large banks in the city of Boston. A brief run of its lending records is extant from the turn of the century, and close examination reveals that very few of Suffolk's loans derived from activities of the sort that White would label as special. For example, 46 percent of the Suffolk's portfolio (by value) consisted of collateral loans, 84 percent of which were granted to brokers and other intermediaries who dealt in the securities markets. These loans were backed by securities that bank examiners regarded as readily marketable. That is, in lending on the basis of this collateral, the Suffolk Bank was not making use of any special informational advantage; it was only accepting securities that were generally perceived by the investment community to be of high quality. The other 54 percent of Suffolk's portfolio consisted of short-term loans based on personal security, but approximately two-thirds of these (by value) were notes purchased on the commercial-paper market. Only 19 percent of the bank's portfolio consisted of notes backed by personal security (one- or two-name paper) that were discounted for the benefit of signatories who were customers of the bank. In other words, only 19 percent of the bank's portfolio consisted of loans for which the bank might perform some special informational role (Lamoreaux 1994, p. 128).


Suffolk, of course, is just one example, but it is an example that I would argue was representative of larger trends in the banking system. Unfortunately, it is impossible to demonstrate this claim by analyzing a large sample of banks' loan portfolios. Very little information of this sort is extant. Instead, it is necessary to approach the problem more indirectly - by thinking about what is known about banks' management structures and practices during this period.

Nineteenth-century banks typically had very lean managerial hierarchies. For example, national banks were governed by a board of directors, one of whose members was elected president. The daily affairs of the bank were generally run by a cashier, who might work alone or, depending on the size of the institution, might be assisted by one or more tellers and clerks, and perhaps a bookkeeper. For most of the century, the cashier was the chief operating officer. Presidents (like other members of the boards of directors) were usually part-time officers. They had other business interests to which they devoted their primary attention (Lamoreaux 1994, pp. 3-4).

This type of managerial structure underwent some changes over time. For example, by the end of the century it was increasingly common, especially at large urban banks like Suffolk, to have presidents who had previously served as cashiers and who devoted all their time to their banks (Lamoreaux 1994, pp. 123-4). But the important point is that, whether the chief of operations was the cashier or the president, he supervised relatively few people. In particular, there was little investment in the nineteenth century in developing the organizational capability to collect and process information about the creditworthiness of borrowers. To the extent that banks had an informational advantage over other financial institutions, it was a personal one that derived from having a chief officer who was well connected locally, had repeated dealings with the same people, and spent time (as A. Levy did) traveling around the community checking up on borrowers.(1) A bank might also gain additional information about potential borrowers by choosing for its directors people who had good knowledge of particular segments of the business community and who were willing to use this knowledge for the benefit of the bank.

The kind of information that nineteenth century banks acquired through their officers was thus local and personal. It derived from the imbeddedness of these men in the communities from which most of the institutions' borrowers were drawn. A lender who was not similarly imbedded did not have access to information of comparable quality. Such a lender might subscribe, for example, to the reports of credit agencies like the R.G. Dun Company, which had corresponding agents, often lawyers, located in communities throughout the country. These agents gossiped with local merchants, kept their ears open, and reported any news that might affect the creditworthiness of potential borrowers. There is reason to believe, however, that for much of the century this kind of information was inferior in quality to that obtained by local bank officers for their institutions. After all, bankers had access to all the same sources of information as the agents of the credit agencies. In addition, they had the direct knowledge that came from their own private dealings with borrowers.

Even in this period, however, there were important limits on the kinds of information that banks were able to collect. For example, it was not generally considered appropriate to ask borrowers for financial statements. Even bankers, therefore, had only impressionistic evidence of their borrowers' net worth. Like White's Achille Levy, they based their lending decisions as much on their assessment of an applicant's character as on precise information about income and liabilities.


This limitation on information collecting would become more important over time. By the end of century, the informational advantage that bankers derived from their local imbeddedness and from their repeat dealings with borrowers was increasingly inadequate (and regarded as so by contemporary observers), especially in the most economically developed and urban areas of the nation.(2) Part of the problem was the trend toward single-name paper that White discusses in his essay. Loans on personal security were more risky than they had been earlier because they were backed only by the wealth of the maker, not the maker plus one or more endorsers approved by the bank. A more important problem for our purposes was the growing tendency for businesses to borrow from more than one financial institution and also to float commercial paper on the market. As a result, it was now more difficult for bankers to get a good sense of a borrower's financial position just from their own repeated dealings with the individual (James 1978, pp. 55-59; Lamoreaux 1994, pp. 89-90).

Banks dealt with this problem by moving to require formal, sometimes audited, financial statements from borrowers. They also began to invest in information-gathering capabilities, creating new credit departments whose business was to keep track of customers' creditworthiness. These developments came relatively late. As White tells us, financial statements were not in common use, even in large banks, until the 1890s; in small banks the delay was much longer. Credit departments were also first organized in the 1890s. As late as 1899, only 10 banks had them, and they were all in New York.

I would like, however, to question whether these investments in the organizational capability to collect information about borrowers really did much good in the sense of allowing banks to recapture their informational advantage over other kinds of financial institutions. AfteR all, any lender could require a financial statement. Moreover, to the extent that borrowers sought loans from multiple institutions and floated commercial paper on the market, banks were not particularly well placed to assess the truthfulness of these statements. As a result, banks ultimately had to depend on external sources of verification such as independent audits, information collected by credit agencies, and (later) tax returns - sources of information that were available on the same terms to other financial institutions.

There was, of course, another possibility Banks could have embraced what Charles Calomiris and others have called "universal banking" and developed long-term relationships with the companies to which they lent funds, taking equity positions in the firms and naming directors to their boards. In that way, they could have gained inside information about income and net worth and also, perhaps, some say over the companies' managements. But U.S. banks did not go this route.(3) Instead, they coped with their growing information problems in a number of alternative ways: by requiring customers to maintain deposits with them worth a certain percentage of their credit line;(4) by keeping borrowers on a short leash and forcing them to renew their loans frequently;(5) and even, as the Suffolk Bank essentially did, by giving up the whole idea of maintaining an informational advantage and engaging in the kinds of lending that did not require special knowledge (for example, lending to stock-market brokers on the security of readily marketable stocks or buying commercial paper on market). Suffolk's strategy, by the way, was not particularly profitable, and the bank's stockholders voluntarily reduced its capital in the early twentieth century. Banks' other coping mechanisms also proved ultimately impracticable as pressures from regulators and competition from other lenders forced them to give up the idea of compensating deposits and to lengthen lending terms in the twentieth century (Lamoreaux 1994, pp. 101-2, pp. 136-7, p. 163).

In the end, therefore, banks' loss of specialness was not so much a result of a fall in the cost of information as of a lack of advantage in collecting the kinds of information needed to assess the creditworthiness of borrowers operating in the geographically wider markets of the twentieth century. Information costs are undoubtedly a part of this story, but not, I think, in the way White originally intended. The story I would tell would be one that emphasized the decline in local lending. Banks were special in the nineteenth century because, unlike most other financial institutions, they were located close to their borrowers. As local lending declined, banks' informational advantage disappeared. Not surprisingly, their market share also dropped as a consequence.

1 Thus William Goddard felt he had to resign from the presidency of the Providence National Bank in Providence, Rhode Island, when he was afflicted with "an obstinate lameness [that] so tethers me to the spot ... that I feel disqualified from seeking elsewhere the information regarding the credit of borrowers, which I regard as of the highest importance to the successful management of a bank" Quoted in Lamoreaux (1994), p. 103.

2 See, for example, Rhodes' Journal of Banking (February 1893, p. 137; April 1893, p. 377; and June 1893, p. 585) and Bankers' Magazine (January 1893, pp. 525-26; August 1898, pp. 286-87; and September 1898, p. 384 and pp. 413-22). See also Cannon (1891), pp. 535-36.

3 It is beyond the scope of this commentary to explore the reasons that U.S. banks did not practice universal banking. For two alternative views, see Calomiris (1995) and Lamoreaux (1994).

4 As one banker wrote in Rhodes' Journal of Banking, "the best paper to accept is that offered by firms or individuals who are in the habit of carrying balances with their bank from whom the accommodation should be obtained. There appears to me no better means to determine the amount of risk a bank incurs than by regulating its loans according to the average balance carried" (May 1893, pp. 486-89). See also Forgan (1920), p. 7.

5 Bankers felt that short-term loans helped to discipline borrowers. As one put it, "borrowers will use the proceeds of loans which they are to repay in a few months more wisely than might be the case if the payment were indefinitely deferred." Moreover, because the information in a borrowers' financial statement reflected current conditions, which were liable to "change radically for the worse" with the passage of time, such a statement was clearly "a basis for short time credit only." Coman (1907); and American Institute of Banking (1916).


American Institute of Banking. Loans and Investment, New York: privately printed, 1916.

Calomiris, Charles W. "The Costs of Rejecting Universal Banking: American Finance in the German Mirror, 1870-1914," Coordination and Information: Historical Perspectives on the Organization of Enterprise, Naomi R. Lamoreaux and Daniel M. G. Raft, eds., University of Chicago Press, 1995, pp. 257-315.

Coman, E. "Requisites of a Goad Loan," Practical Problems in Banking and Currency, Walter Henry Hull, ed., Macmillan, 1907, pp. 66-77.

Cannon, James G. An Ideal Bank, privately printed, 1891.

Forgan, James B. A Good Note, Chicago: privately printed, 1920.

James, John A. Money and Capital Markets in Postbellum America, Princeton University Press, 1978.

Lamoreaux, Naomi R. Insider Lending: Banks, Personal Connections, and Economic Development in Industrial New England, Cambridge University Press, 1994.

Naomi R. Lamoreaux is a professor of economics and history at the University of California-Los Angeles end a research associate at the National Bureau of Economic Research.
COPYRIGHT 1998 Federal Reserve Bank of St. Louis
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1998 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:comments on the article of R. Alton Gilbert, in this issue, p. 121
Author:James, John A.
Publication:Federal Reserve Bank of St. Louis Review
Date:May 1, 1998
Previous Article:Did the Feds's founding improve the efficiency of the U.S. payments system?
Next Article:American banks during the Great Depression: a new research agenda.

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters