New York Clearing House Association: The Crisis of 1893

The crisis of 1893, unlike previous crises, began in the interior of the country—with smaller western banks exposed to the railroad industry (Wicker 2000, 52). In the weeks leading up to June 17, 1893, New York banks’ reserves contracted sharply from $134 million to $110 million; surplus reserves—reserves held in excess of the reserve requirement—fell from $24.6 million to $8.7 million (Sprague 1910, 168). Sprague attributes the fall in reserves to the “failure and suspension of large numbers of banks, both state and national, and of private bankers in the West and South” (1910, 168).

By June 14, 1893, the NYCH began putting in place the infrastructure to issue CLCs, at a far earlier stage of the crisis than in previous crises; on June 14, the Clearing House Committee met and called a meeting of the association for June 15 (Sprague 1910, 409). On June 15, the NYCH Association met, and with 58 of 65 banks present, adopted a resolution to appoint a Loan Committee to begin issuing the first CLCs on June 17 (Sprague 1910, 410). See Key Design Decision 2 “Legal Authority” for a description of these committees.

On June 16, the NYCH said that the purpose of issuing CLCs was a:

The NYCH was a private institution, and its emergency liquidity operations developed from its original operations as a place for settlement between banks. The NYCH had an internal process for authorizing CLC issuance. In a time of stress or panic, the NYCH would appoint a Loan Committee, which would take in collateral and issue CLCs to member banks (Gilpin and Wallace 1904, 18).

The National Bank Act of 1864 sought to discourage state-chartered banks from issuing bank notes and charged a tax of 10% of the value of any such notes each year (Timberlake 1984, 6). Contemporary sources and scholars have argued that some of the forms of debt issued to replace currency during crises of the National Banking Era—including “certified checks, certificates of deposit, cashier’s checks in round amount”—should have been subject to the 10% tax (Hepburn 1903, 374).

However, contemporary sources and scholars tend to agree that CLCs used entirely for interbank transactions did not break this law because they circulated only among the NYCH members. William J. Gilpin, assistant manager of the NYCH during the 1893 crisis, stresses in his book on the NYCH that “the certificates were in no sense a circulating medium, being confined to Clearing House uses” (Gilpin and Wallace 1904, 18).

In the 1893 and 1907 banking crises, some clearinghouses outside New York issued CLCs in smaller denominations and encouraged their use as currency, raising more directly the question of their legality under the National Banking Act. Cannon, in his report to the National Monetary Commission, defends the use of CLCs in general but hints that their use in some states could have subjected them to the 10% tax:

. . . their sole function consisted in discharging the single obligation of the clearing house. An attempt on the part of a bank to use them otherwise would have incurred a fine and other penalties, provided in the rule governing such associations. In no instance, except those mentioned in the South, were they designated to serve as money, and nowhere else did they circulate as money. (Cannon 1910, 116)

Banks also typically used certified checks in place of currency during crises when they had partially suspended the cash convertibility of deposits. Starting in 1873, the NYCH guaranteed certified checks issued by its members during crises. Such checks were “quasi-currency” and circulated at a discount to currency (Timberlake 1984, 5). This “currency premium” became a measure of market confidence in banks during crises. Gorton and Tallman (2016) says that certified checks were not considered “money” and were not subject to the 10% tax, but that this was “essentially a fiction” (Gorton and Tallman 2016, 27).

Suspension of deposit convertibility into cash was also effectively illegal because demand deposit contracts gave depositors the right to withdraw cash any time a bank was open. However, those contracts were never legally enforced, as suspension was seen as a necessary tool to calm banking panics at the time (Gorton and Tallman 2018, 181).

For a broader discussion of the legality of CLC issuance, see (Fulmer 2022b).

In other panics, the NYCH deployed a number of tools to quell the panic: (1) anonymous temporary lending, (2) suspending publication of weekly bank balance sheet data, (3) examining member banks (like stress testing), (4) depositor bail-ins of member banks, and (5) suspending cash convertibility for member banks (Gorton and Tallman 2018, 7). Although during the Panic of 1873 and the Panic of 1907, the NYCH suspended cash payments when CLCs were first issued, during the Panic of 1893, the NYCH did not suspend cash payments until August 3, 1893, six weeks after approving CLC issuance (Wicker 2000, 53). The burden of the suspension of cash payments fell heavily on interior banks, while in the Panic of 1873, the NY banks were able to maintain convertibility requests from interior banks, in part due to reserve pooling (Wicker 2000, 77–78).

The NYCH examined two banks during the panic, Madison Square Bank and National Bank of Deposit, neither of which was a member of the clearinghouse (Gorton and Tallman 2016, 72). Madison Square Bank approached the NYCH with its books in an attempt to prove its soundness. National Bank of Deposit was examined after an NYCH member bank refused to act as a clearinghouse agent on its behalf. After examining, the NYCH recommended liquidation (Gorton and Tallman 2016, 72).

Throughout the crisis, the NYCH changed its disclosure guidelines, suppressing the publication of weekly member bank statements and publishing only aggregate statistics (NYCH 1890, 41).

US Department of the Treasury actions during the crisis to expand the supply of money helped to alleviate the contraction. Excess Treasury payments expanded the supply of money by $9 million in August. Combined with gold imports and silver purchases, money in circulation increased by $17 million in July and nearly $70 million in August (Sprague 1910, 184).

The NYCH’s administration comprised a president, secretary, manager, assistant manager, and five standing committees (Paton 1895, 600).^FNThe five standing committees were: the Clearing House Committee, the Conference Committee, the Nominating Committee, the Committee on Admissions, and the Arbitration Committee. Each was staffed with five members, each an officer of a NYCH member bank (Paton 1895, 600). In times of stress or pressure, the association would meet and appoint a Loan Committee, containing five bank officer members and the president of the NYCH as an ex officio member (Paton 1895, 608). The association itself created the committee, authorized CLC issuance, and specified the interest rate and accepted collateral. The Loan Committee took in and valued collateral, issued CLCs, collected and disbursed interest payments, and, upon initiation by a borrowing bank, processed CLC redemption (Paton 1895, 608).

The CLC committee met daily throughout the crisis, beginning June 15 (Sprague 1910, 410). Gilpin, who was appointed to the role of assistant manager of the NYCH in 1892, supervised bank withdrawal of gold coin deposits and the issuance of CLCs (Gilpin and Wallace 1904, 35). The chairman (later called the president) of the NYCH during the crisis was George G. Williams, president of the Chemical National Bank (Gilpin and Wallace 1904, 47).

On June 15, the NYCH Association met, authorized the issuance of CLCs, and appointed the Loan committee to value collateral and oversee issuance (Sprague 1910, 410).

To receive a CLC, a borrowing member bank would approach the loan committee with suitable collateral, and after applying a 25% haircut, would receive a certificate (OCC 1893, 16; Paton 1895, 608). The CLC could be used in settlement of payments with any other NYCH member bank; the accepting bank was paid 6% interest on the CLC as long as the CLC was outstanding (Sprague 1910, 410). Interest was paid by the borrowing bank to the Loan Committee, which then disbursed interest payments to accepting banks (Hoag 2016, 310). Collateral was held in the vaults of Chase National Bank and First National Bank, as the NYCH itself lacked the capacity to secure the collateral (Sprague 1910, 412). Should a bank default on a CLC, the Loan Committee was authorized to liquidate securities deposited as collateral to satisfy the loan (Gilpin and Wallace 1904, 18).

For more on the function of CLCs, see (Fulmer 2022b).

CLCs were issued only to member banks of the clearinghouse for temporarily settling balances with other member banks (Paton 1895, 602). By 1893, the NYCH had 65 members (Gilpin and Wallace 1904, 33). In total, 38 banks took out CLCs; 27 did not (OCC 1908, 65).

CLCs, as Sprague says in his report to the National Monetary Commission, were “temporary loans made by the banks associated together as a clearing-house association to the members of such association” (Sprague 1910, 406). Moreover, “their sole function consisted in discharging the single obligation at the clearing house” (Sprague 1910, 407). Should a bank default on a CLC, the Loan Committee was authorized to liquidate securities deposited as collateral to satisfy the loan (Gilpin and Wallace 1904, 18).

While the CLCs were, in one sense, the obligation of the borrowing banks, they were also jointly guaranteed by the association as a whole. Cannon describes CLCs as the nonborrowing member banks taking on the “joint risk of the [borrowing] bank in proportion to the amounts of their respective capitals” (1910, 241). Therefore, if liquidated collateral could not fully redeem a CLC, members jointly shared the remaining loss. See (Fulmer 2022b) for a more complete discussion of the joint nature of CLCs.

NYCH member banks approached the Loan Committee when they needed certificates for settlement; thus, the overall size of the CLC issuance was limited only by member bank demand.

The first round of CLCs was issued on June 17,^FNThere is some disagreement: Cannon states that the first round of CLCs was issued on June 21, 1893 (1910, 98). with a value totaling $2.6 million (Sprague 1910, 410). The last CLCs were issued on September 6 (Cannon 1910, 98). In total, $41.9 million CLCs were issued; the greatest amount outstanding was on August 29, at $38.3 million (Sprague 1910, 411). At peak, outstanding CLCs constituted 10.2% of NYCH banks’ deposits (Sprague 1910, 411).

By July 1893, 78% of balances were settled with CLCs; in August, this value increased to 95% (Sprague 1910, 182) (see Figure 3). By October, certificates were no longer used in settlement balances, and by November 1, all CLCs had been redeemed (Cannon 1910, 98). This high percentage indicates wide usage among member banks; in other panics, only a few banks used CLCs (Hoffner 2022).

Source: Sprague 1910, 411.

Research did not reveal individual participation limits. Individual issuance was naturally limited by the eligible collateral each bank had on hand.

The CLCs bore interest at a rate of 6% (Sprague 1910, 410). In total, $535,513 was paid as interest on issued CLCs (Sprague 1910, 412). Although in other panics the NYCH charged administration fees, there is no indication that the NYCH charged fees in the panic of 1893 (Fulmer 2022a). In a 1910 report to the National Monetary Commission, James Graham Cannon^FNJames Graham Cannon was vice president of the Fourth National Bank and an administrator of the NYCH (Gilpin and Wallace 1904, 40). explained that the interest rate on CLCs encouraged timely repayment:

The aim is to fix the rates sufficiently high to ensure the retirement of the certificates as soon as the emergency which called them forth has passed by. As a rule they are retired by the banks, which take them out as soon as they have obtained sufficient cash to meet their daily obligations (Cannon 1910, 79).

To receive CLCs, a bank was required to deposit collateral of “its bills receivable, or other securities”—including stocks and bonds, at which point the NYCH Loan Committee issued CLCs (OCC 1893, 16; Paton 1895, 608).^FNIn prior crises, the NYCH accepted a different range of collateral; the range of eligible collateral expanded over time. See Paton (1895, 608) for more. A bank could borrow at most 75% the value of its posted collateral (OCC 1893, 16).

In total, the NYCH received $56 million of collateral—$40 million in bills receivable and $16 million in stocks and bonds (Sprague 1910, 411).^FNGiven the total sum of CLCs issued the implied haircut is 74.1%.

Collateral was held in the vaults of Chase National Bank and First National Bank, as the NYCH itself lacked the capacity to secure the collateral (Sprague 1910, 412).

CLCs were issued without a maturity date and, for purposes of settlement between banks, could circulate freely between NYCH members until redemption. The first cancellation of a CLC occurred on July 6, 1893 (Sprague 1910, 410).

In the case of the Panic of 1893, the last of the CLCs were redeemed on November 1, more than four months after the first issuance (Cannon 1910, 98).

Member banks were required to accept CLCs in daily settlement, as a condition for membership in the association (see, for example, Cannon 1910, 79). See (Fulmer 2022b) for a more complete discussion of the NYCH membership requirements.

As a private organization, the NYCH did not consider monetary policy impact in its issuance of CLCs, as the clearinghouse was a private association of banks. However, Cannon describes how CLC issuance increased the money supply in periods of high demand:

In 1873, the NYCH took the extra step of pooling reserves—essentially jointly sharing reserves to shore up weaker banks (Fulmer 2022a). However, in 1893, banks were less willing to use this extraordinary measure:

Sprague indicates that reserve pooling would have prevented the crisis from deepening to a point that necessitated the suspension of cash convertibility:

Sprague notes that New York banks were reluctant to employ reserve pooling in the crisis of 1893 partly because some New York banks were more exposed to interior banks than others; unexposed banks were unsympathetic to the suffering of exposed banks (Sprague 1910, 185–86).

Other cities’ clearinghouses issued CLCs in response to the crisis. These cities included Philadelphia, Boston, Baltimore, Pittsburgh, New Orleans, Cincinnati, Buffalo, Detroit, Atlanta, Birmingham, and Richmond (Cannon 1910, 98–112). Many of these cities issued small-denomination CLCs that to some extent circulated outside the clearinghouses as currency (see Cannon 1910, 112). Issuance was far more widespread than in previous crises: in the 1873 crisis, New York and Philadelphia clearinghouses issued CLCs, while in the 1890–91 crisis, New York, Boston, and Philadelphia issued CLCs (OCC 1893, 16). While the Chicago clearinghouse authorized CLCs, the banks did not take out any CLCs, likely due to stigma, which might have contributed to the deeper loan contraction there (Sprague 1910, 209). See Figure 4.

On June 15, 1893, the New York Times reported that the NYCH Clearing House Committee had met to discuss the issuance of CLCs, and though the NYCH hadn’t issued a formal statement, “it is understood, however, that the committeemen are in favor of an issue of certificates” (NYT 1893b). The article further stated that the NYCH Association would meet the following day, June 16, likely to authorize CLC issuance (NYT 1893b). The June 17, 1893, issue of the Commercial & Financial Chronicle included a discussion of the NYCH’s decision to begin offering CLCs (Commercial and Financial Chronicle 1893, 1000). On November 3, 1893, the New York Times reported that the last of the CLCs had been redeemed and that the NYCH Association published a report on CLC issuance on November 2, 1893, detailing its operations throughout the crisis (NYT 1893a).

On October 31, 1893, the Loan Committee submitted a report to the association, including aggregate statistics on the operations of the committee (Sprague 1910, 409). The text of this report was published in the New York Times on November 3, 1893 (NYT 1893a). Attached to the report, the committee included “full and complete statistics of the transactions with each bank” (Sprague 1910, 412).

However, throughout the crisis, the NYCH changed its disclosure guidelines, suppressing the publication of weekly member bank statements and publishing only aggregate statistics (NYCH 1890, 41).

In Cannon’s report to the National Monetary Commission, he describes the stigma associated with the CLCs: “Some banks regard their use of clearing-house loan certificates as a reflection upon their standing, and hence refuse to apply for them unless driven to it by sheer necessity” (Cannon 1910, 79).

However, in general, prior to the Panic of 1907, NYCH member banks viewed the use of CLCs as a “patriotic movement in which all the banks should engage, both for the purpose of assisting their fellow-members and for the welfare of the community as a whole” (Cannon 1910, 79). High usage of CLCs throughout the crisis indicates wide subscription by member banks, unlike in other crises. By comparison, in the Panic of 1907, only 60% of members took out CLCs, indicating that their usage might have become stigmatized (see Runkel 2022).

On June 14, 1893, as in previous crises, the NYCH suspended publication of weekly member bank statements and opted to “furnish the public with the totals and changes only” (NYCH 1890, 41). Delaying or suspending publication of bank statements prevented bank runs from worsening and prevented the public from identifying weak banks (Gorton and Tallman 2018, 44). On November 3, 1893, the New York Times published a Loan Committee report on CLC issuance throughout the crisis (NYT 1893a). Attached to the report, the committee included “full and complete statistics of the transactions with each bank” (Sprague 1910, 412). On November 4, 1893, the NYCH resumed the publication of weekly bank statements (NYCH 1890, 154).

The last CLCs were issued on September 6 (Cannon 1910, 98). By October, certificates were no longer used in settlement balances, and by November 1, all CLCs had been redeemed (Cannon 1910, 98). Interest rates were set high enough to encourage timely repayment; See Key Design Decision No. 12, Loan Duration, for a more detailed discussion.