Outwardly and according to its balance sheet, the [Knickerbocker] Trust Company was flourishing.

—Herbert L. Satterlee, J. P. Morgan’s son‐in‐law and biographer¹

An Imposing Edifice

The Knickerbocker Trust Company once stood at the northwest corner of Fifth Avenue and 34th Street, in a transitional neighborhood between New York’s downtown business district and its uptown residential grandeur.^2,3 On the opposite corner from the Knickerbocker towered the red sandstone palace of the Waldorf‐Astoria, New York’s most fashionable and grand hotel, placing the Knickerbocker literally in the shadow of the city’s social, business, and political hub. As such, the Knickerbocker’s designers hoped the building itself would be commensurate with its genteel surroundings. Covered in Vermont marble and fronted by four 17‐ton Corinthian columns, the Knickerbocker presented, according to a contemporary reviewer, “a beautiful example of Grecian architecture, treated in the most refined way with every provision known to modern building construction for meeting the demands of the company and the service of its patrons. The structure must be considered a distinct addition to the architectural features of that part of the city.”⁴

The Knickerbocker’s architects, McKim, Mead & White, had designed an elegant and functional four‐story structure that was intended to convey the institution’s strength and sobriety. Inside its main gates, white Norwegian marble contrasted brilliantly with interior bronze detailing and mahogany woodwork. The central banking room reached nearly three stories high and had eight adjoining rooms for ancillary banking activities. An alternate entrance on 34th Street led to second, third, and fourth floors, each with 5,000 square feet of floor space and four executive offices. In the basement, the massive safety deposit vault contained 2,000 boxes, and its outer vault door weighed nearly nine tons—its hinges alone were 3,700 pounds. Elsewhere in the building was a 6,000‐square‐foot employee dining hall with a full kitchen.⁵

The Knickerbocker building thus captured the prevailing ethos of early‐twentieth‐century bank construction, as described in Banker’s Magazine, a contemporary industry periodical:

[T]he public expects a bank or trust company to occupy well‐furnished quarters, such that the cost of good equipment is more than repaid in advertising value. The lobby should be provided with all possible conveniences for customers, including writing desks supplied with good ink, clean pens and blotters, comfortable chairs, etc. The larger companies often provide separate reading and lounging rooms for customers, with current newspapers and magazines, writing desks and other conveniences, and other committee‐rooms for the use of customers who wish to meet or arrange details of business. In many cases there are special quarters for women, equipped with numerous conveniences, and in charge of a matron who looks after the comfort of patrons.⁶

The Insurgency by Trust Companies

As a trust company, the Knickerbocker represented a relatively new form of financial intermediary. Originally organized in the mid‐nineteenth century to handle various financial tasks for private estates and corporations, the sphere of activity for trust companies gradually expanded to offer services little different from those of traditional banks.^a As historian Vincent Carosso has explained:

Beginning in the 1890s, trust companies took on most of the functions of both commercial and private banks. They accepted deposits; made loans; participated extensively in reorganizing railroads and consolidating industrial corporations; acted as trustees, underwriters, and distributors of new securities; and served as the depositories of stocks, bonds, and titles. Corporations regularly appointed them as registrars or fiscal and transfer agents. Very often they also owned and managed real estate.⁷

Despite their functional similarities to national and state banks, trust companies were generally less tightly regulated. They were permitted, for instance, to hold a wider variety of assets; unlike national banks, trust companies could own stock equity directly. Also, unique among large financial institutions, trusts were not required to hold reserves against deposits before 1906; in that year, New York State required that trust companies hold 15 percent of deposits as reserves, although only a third of the reserves had to be held in cash—in contrast to New York State regulations, national banks in the city had to meet a federal government minimum reserve of 25 percent. This meant that trust companies could earn a higher return on their assets compared to banks, and thus could pay higher interest rates. Accordingly, the higher interest rates attracted deposits, and the trust companies’ deposits grew rapidly. In 1906, the assets of all trust companies in New York City, approximated the assets of all national banks in the city, and exceeded the assets of all state banks there. According to economists Jon Moen and Ellis Tallman, “In the ten years ending in 1907, trust company assets in New York State had grown 244 percent (from $396.7 million to $1.364 billion) in comparison to 97 percent (from $915.2 million to $1.8 billion) for those of national banks, and 82 percent (from $297 million to $541 million) for state banks in New York.”⁸

Trust companies were “shadow banks” that originated on the periphery of the formally regulated financial system and exploited their fringe status to gain a growing share of the financial services market. They owed their growth to a different business model, which exposed the banks to “an unprecedented amount of competition,” according to Professor Sprague.⁹

First, they reached out to a growing segment of depositors, especially the middle class. In a booming economy, individuals could save more of their take‐home pay. Throughout the nineteenth century, banks had generally not actively marketed their services to depositors and instead focused on serving merchants, large corporations, and correspondent banks. Indeed, some institutions, such as J.P. Morgan & Company (a private bank) accepted no deposits from the general public—you had to be recommended to become a depositor, as if joining a selective club. But trust companies welcomed individual depositors and promoted themselves more aggressively than banks through advertising and ethnic affiliation, by locating in convenient locations to individual depositors, and by paying interest on deposits, a practice that conservative bankers frowned upon. Unlike banks, trust companies could open branch offices in neighborhoods convenient to their clientele. The Knickerbocker, for instance, operated three branches in addition to its main office. Trust companies were exemplars of the growing democratization of finance in the sense of providing increased access to financial services for segments of the public not formerly served.

Second, in order to pay interest on deposits, trust companies held lower cash reserves against deposits and invested in real estate, stocks, and other longer‐term assets that yielded higher returns. Moen and Tallman estimated that 92.2 percent of trust company assets in New York earned interest, in comparison to 70 percent at nationally and state‐chartered banks.¹⁰ New York State imposed lighter reserve regulations on trust companies. The practice of investing in longer‐term assets meant that the trust companies’ loan portfolio was less liquid than national banks. In short, trust companies were riskier (but probably yielded a higher return on investment) than national banks.

Third, trust companies in New York distanced themselves from the mainstream banking system. They engaged much less than banks in the payments system (i.e., moving money around in the form of cash, checks, bills of exchange, etc.). Classic trust deposits (not bankers’ deposits) tended to be held on a long‐term basis, in trust for beneficiaries, and were less subject to transactional inflows and outflows. Moen and Tallman estimated that the volume of transactions services at New York trust companies was only 7 percent that of national banks. Whereas trust companies accepted deposits from other banks (a practice called correspondent banking), such deposits accounted for less than a sixth of total deposits.¹¹

Estrangement from the Clearing House

At the time of the Panic of 1907, only 3¹² of the 38¹³ Manhattan trust companies were affiliates of the New York Clearing House. Four years earlier, facing steadfast insistence by the NYCH to increase reserves or leave, most of the trust companies decamped. Their absence from the NYCH meant that trust companies could not look to it as a lender of last resort in a panic. This split fueled later charges by Senator Robert La Follette, Representative Arsene Pujo, and investigator Samuel Untermyer that the Panic of 1907 sprang from vengeful discipline of the trust companies by the banks.¹⁴

Yet the trust companies themselves were hardly homogeneous. They varied considerably by history and business model. Younger firms pursued a retail‐oriented strategy; older firms pursued a wholesale‐oriented strategy. The oldest firms’ executives were well‐known to J.P. Morgan and his circle. The smaller and newer trust companies were generally outsiders to Morgan’s social and business circles.

The Geography of Trust

Historian Bradley Hansen noted distinctive differences between trust companies that were located downtown (in the vicinity of Wall Street) and those that were uptown, which was the scene of more retail and residential activity. Figure 9.1 gives a map showing the locations of 38 Manhattan‐based trust companies in 1907: what stands out is the tight cluster of 28 of them in the financial district, with the other 10 loosely dispersed north toward “uptown.”

Hansen said that the uptown trusts had “large numbers of relatively small deposits… . Downtown firms recovered quickly [from the panic]; uptown firms did not recover.”¹⁵ Caroline Fohlin and Zikhun Liu noted that seven trust companies were closely affiliated with J.P. Morgan, Baker’s First National Bank, Stillman’s City National Bank, or the National Bank of Commerce. Potentially the affiliated trust companies could look to their bank affiliates for lender‐of‐last‐resort assistance in the event of panics. Other institutions noted earlier were affiliated with the Heinze‐Morse circle. As Hansen concluded, “on the eve of the Panic, New York City trust companies were far from being a homogeneous group.”¹⁶

Our own analysis of data about the trust companies (see the Technical Appendix after Chapter 24) supports the relevance of business model differences, and associations with the Heinze‐Morse circle in explaining deposit losses. And the results lend but weak support to the hypothesis that market power or links to the financial elite (Morgan and his circle, or the NYCH) resulted in better deposit results.

Whereas diversity of players in a competitive field should be no surprise, the diversity portended a fatal flaw: collective action among the trust companies to save themselves in the event of a panic would be significantly more difficult than for the bank members of the NYCH.

The Problem of Dependency

A final concern in the pre‐panic period was the dependence of New York City financial institutions on correspondent bank deposits. In modern terms, these deposits might be called “hot money” owing to their tendency to be withdrawn quickly in the event of an episode of systemic instability. In addition, banks in agricultural regions of the interior tended to withdraw deposits to fund credits during the annual planting/harvest cycle. The recipients of bankers’ deposits, New York banks and trust companies, tended to invest such deposits in demand loans (loans with no fixed maturity that were payable upon demand), and especially in collateralized margin loans to brokers on the stock exchange (“call loans”) because of the higher interest rates and liquidity of those loans. And a very aggressive trust company might use bankers’ deposits to fund investments in fixed term loans such as real estate mortgages.

In the event of a panicked withdrawal of bankers’ deposits from trust companies, the trust companies would call in their brokers’ loans, forcing market participants to sell stocks, depressing stock prices, and putting downward pressure on all speculative lending. Thus, a sudden exit by the trust companies would present the banks with the unpalatable choice of stepping in to provide liquidity to the stock market (and thus protect their own call loan business) or watching the market crash. As Sprague put it, “If, for any reason, it should become necessary for the trust companies to contract their banking operations, it would obviously be necessary for the banks to shoulder the burden in order to save the local situation.”¹⁷

What made this panic scenario troubling was the fact that since 1897, trust companies had claimed the lion’s share of growth in bankers’ deposits in New York City, as shown in Figure 9.2.

Bankers’ reserves at New York financial institutions grew from 1897 to 1906, at a 7.6 percent compound annual rate, well in excess of the growth of GDP per capita of 3.3 percent over the same period. Handling bankers’ deposits was the growth vehicle for New York financial firms over those years.

As importantly, New York City trust companies had seized the dominant share of deposits, in comparison to national banks in New York, Chicago, and St. Louis, the principal reserve cities in the United States. Examination of the growth in trust company bankers’ deposits reveals that deposits from state banks constituted a significant portion of the growth over the years 1897–1906.

A final detail warrants consideration: the largest banks and trust companies gained the dominant share of this growth in bankers’ deposits, suggesting a “winner take all” dynamic in the market. The data for national banks give some indication of what was true for trust companies: of New York City’s 38 national banks, six of them accounted for three quarters of such deposits.¹⁸ The same banks accounted for more than half of individual deposits, and nearly 60 percent of national bank loans. As Sprague pointed out, the significance of this concentration was that these six banks had greater exposure to a financial crisis, and greater power for mobilizing collective action to quell it.¹⁹

A Pillar of the Financial Community

By 1907, the Knickerbocker Trust Company had become one of the largest and most successful trust companies in the country. Originally organized in 1884 by Fred Eldridge, who happened to be an old friend and classmate of J. Pierpont Morgan,²⁰ the Knickerbocker had grown its deposits in recent years by nearly 40 percent. By 1907, the Knickerbocker reported nearly $65 million in deposits, making it the second‐largest trust company in New York City by deposits, with nearly 18,000 depositors. The firm acted chiefly in the traditional roles of trust companies as an executor, administrator, guardian, receiver, registrar, transfer, and financial agent for states, cities, railroads, corporations, trusts, and estates. But, like other trust companies, the firm also offered interest on time deposits and received deposits subject to demand checks.

The architecture of the Knickerbocker’s new building at the corner of Fifth Avenue and 27th Street, completed in 1906, justly reflected its prominence. The company also maintained a large office downtown near Wall Street and two remote branches in the Bronx and Harlem—at this time, branches were an innovation among the trust companies.²¹

Adding to the luster of the Knickerbocker was its president, Charles T. Barney, a leading figure in New York’s financial and social circles. Born in Cleveland in 1851, Barney graduated from Williams College in 1870 and moved to New York to make a career in finance. Not nearly of the same standing as J. P. Morgan or the other heads of large national banks, Barney was still well known owing to his membership in 15 social clubs, “special membership” in an NYSE firm, and his membership on the boards of directors of 32 companies, including Trust Company of America.²² Notably, Barney was a director in Charles Morse’s American Ice Company. Also, Barney’s membership on the boards of directors of the National Bank of North America and New Amsterdam National Bank were links to the chain of banks controlled by Charles Morse. Thanks to his social connections (he was the son‐in‐law of the financier William C. Whitney), Barney had been able to draw large accounts to the Knickerbocker from railroads, banks, and brokerage houses. During the decade of his leadership, the Knickerbocker had multiplied its deposits by 6, its surplus by 5, and its dividends by a factor of 10. “Outwardly and according to its balance sheet, the Trust Company was flourishing,” wrote Herbert Satterlee, J. P. Morgan’s son‐in‐law and biographer.²³ In short, the Knickerbocker Trust Company stood prominently in the New York and even national financial arenas.

So, as the events of October 14–15 on Wall Street unfolded, few would have suspected that the esteemed Knickerbocker Trust Company could ever be drawn into the morass. The firm’s size, the apparent strength of its deposits, the social standing of its president, and even its new building in midtown gave the outward appearance of solidity and security. Of course, in hindsight, some found clues to what would unfold in the following days. “Mr. Barney had a very fine board of directors [at the Knickerbocker],” remarked Satterlee, “but they knew very little of a large part of the business of the Trust Company which Mr. Barney kept ‘under his hat.’ He ran his company with but few board meetings and with scant reports of his operations to his subordinates or executive committee.”²⁴ Barney even once told a newspaper reporter that he did not “propose to waste his time answering questions asked by Directors who might try to keep in touch with the daily details of the company’s business.”²⁵ Moreover, Satterlee added, the company’s “list of depositors contained a very large proportion of people of small means and those whose knowledge of banking was so slight that they would be apt to be frightened at the first sign of trouble.”²⁶ Even so, the public at first remained ignorant of the degree of Barney’s connections to the turmoil.

On Monday morning, October 21, New York City’s bankers were still unsettled from the financial tremors triggered by the activities of F. Augustus Heinze and Charles W. Morse the week before. True panic had not yet set in, but the sudden collapse of the brokerage firm Gross & Kleeberg and the ensuing runs at the National Bank of North America, the New Amsterdam National Bank, and the Mercantile National Bank created an atmosphere of high tension. The quick‐witted intervention by the New York Clearing House (NYCH) had nonetheless led many to believe that the problems had been contained. Professor Sprague noted that “there had been nothing in the nature of a crisis during the week the clearing house was putting its affairs in order.”²⁷ The apparent containment of the crisis occurred because as depositors pulled their funds from institutions controlled by Morse and Heinze, they redeposited them in other New York City banks, thereby mitigating the possibility of massive runs on other national banks. Later scholars Charles Calomiris and Gary Gorton found that “the Panic of 1907 is practically a non‐event from the standpoint of national bank failures.”²⁸ The system had received a shock, but a major banking panic had not been unleashed.

At 1 a.m. on Monday, October 21, the NYCH committee concluded a meeting at which they agreed yet again to make further provisions for the payment of debit balances for the Mercantile, North America, and New Amsterdam banks. The Wall Street Journal reported that the NYCH had committed “to extend all necessary aid to all banks in any way involved in the unsavory transactions of United Copper, thus avoiding the possibility of a “rich men’s run” on the banks.”²⁹ Meanwhile, the new president of the Bank of North America, William F. Havemeyer, who had replaced Charles Morse, attempted to calm depositors’ fears, claiming the bank’s board had made headway in repaying $1.3 million in loans previously made to its own directors—$900,000 had already been repaid; the remaining $400,000 was in loans to Charles Morse. “We are busy trying to get out of debt and have little time to talk,” Havemeyer said. “We are raising money and will keep on doing so. Affairs look brighter all the time.”³⁰