NOMISMA SCHOLA
Read the textbooks TextbooksRoutesColumnFilms & narration
Go deeper by course Courses
Verify with the coins CoinsGlossarySourcesPractice
Think with the patterns Patterns
About the school Start hereFacultyContactThe discipline
Admissions Sign in

Language
SOURCE ROOM

How Did Pre-Fed Banking Panics End?(NBER Working Paper 22036)

Gary Gorton, Ellis W. Tallman — National Bureau of Economic Research

Claims Supported by This Source

  1. During severe panics in the National Banking Era (1863–1914), banks suspended convertibility, private clearing houses shielded information on individual banks, acted as a single entity by issuing loan certificates (joint liability of members), and a market for certified checks opened, creating a currency premium (the price of the clearing house's default risk). When the premium returned to 0, confidence was restored (Abstract).
  2. Clearing house responses (anonymous lending, shielding individual bank information, special examinations, rescue of members) correspond to 2007–08 policies (TAF, TSLF, PDCF, ban on short selling, stress tests, rescue of Bear and AIG) (Section 1).
  3. The main reason for the refusal to support Knickerbocker on October 21 was that the trust company was not a member of the clearing house. Sprague and Wicker view this as a serious error by the clearing house. The Trust Company of America had withdrawals of $500,000 on October 22; a state inspection revealed no loans to Morse and loans of $175,000 to Barney, and J.P. Morgan & Co. lent $1,000,000, but there were withdrawals of $13,000,000 the next day and $9,000,000 on the 24th (Wicker 2000). The Secretary of the Treasury announced on October 24, "I have directed the deposit of $25,000,000 in this city," later increased to $35,000,000. On November 2, the trust companies contributed an additional $15,000,000, and on November 4, $25,000,000 of the trust companies' funds were distributed, and the run subsided (Section 4, pp. 25–26).
  4. Table 3 (Dates of events in New York): Panic of 1907 begins October 21; suspension of individual bank information, first issue of loan certificates, and suspension of convertibility all on October 26; currency premium begins October 31; premium at 0 on December 28; convertibility resumes January 1, 1908; last certificate issue January 30; resumption of individual bank information February 8; final redemption March 28.
  5. Table 5 (Weekly, October 19, 1907–February 1, 1908): Currency premium low/high, call money rate low/high, cumulative net gold imports for New York City, outstanding clearing house loan certificates, reserve deficit. Call money high of 125% for week of October 26, 75% for week of November 2. Premium high of 4% for weeks of November 9 and 16. Certificates outstanding peaked at $87,865,000 for week of December 21. Cumulative gold reached $97,763,000 on January 4, 1908, and $103,095,000 on February 1. Reserve deficit was largest at $54,150,000 for week of November 23.
  6. In 1907, the correlation between the premium and cumulative gold imports was -0.89 (November 2 to January 4). By the week of November 29, cumulative gold imports exceeded $55,000,000. The premium did not fall to 0 until reserve inflows became strongly positive in January 1908 (Section 7E).
  7. Appendix A: Of the 53 clearing house members on October 19, 1907, 4 disappeared from the rolls during the crisis, with 2 counted as failures (Mechanics and Traders reopened as Union Bank of Brooklyn; National Bank of North America paid creditors in full with interest). Wall Street Journal, October 16, 1908: Depositors in the 15 institutions (3 national, 8 state, 4 trust) that closed in New York and Brooklyn would not suffer losses.
  8. Gilpin–Wallace 1904: Since 1860, $168,774,000 in loan certificates had been issued and redeemed without the loss of a single dollar (Appendix).

How to read the verification levels. "Read Through" means the source was read in its entirety. "Bibliography & Abstract Read" means the bibliographic information and abstract were confirmed. "Via Course Material" means the source was referenced through the textbook or course text. The "Text Hash" is the SHA-256 of the main text from the publisher's page, excluding tags and whitespace, used to detect if the source has been modified.