Abstract
Using an extensive high-frequency data set, we investigate the transmission of financial crisis specifically focusing on the Panic of 1907, the final severe panic of the National Banking Era (1863–1913). We trace the transmission of the crisis from New York City trust companies to the New York City national banks through direct and indirect interconnections. Trust companies held cash balances at national banks and these balances were liquidated as trust companies suffered depositor runs. Secondly, trust companies and national banks were notable creditors to the New York Stock Exchange; when trusts were suffering runs, the call loan market on the stock exchange seized. The crisis spread to the interior banks after the New York Clearing House banks restricted the convertibility of deposits into cash. Bond returns were sharply negative in the 2 weeks following the suspension. The suspension of convertibility produced a currency premium, which in turn attracted gold imports from Europe. The New York Clearing House had only limited capability to fight the panic through its use of clearing house loan certificates. The gold imports ultimately restored liquidity to financial markets.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Rich (1989) emphasizes the culpability of the USA as the source of worldwide financial distress in 1907 by draining gold away from key European financial markets. It was considered a truly worldwide panic. The shipment of gold out of the European money centers (London, Paris, and Berlin) and toward New York City in November and December 1907 transmitted the panic from the USA to Europe.
Gorton and Tallman (2016) provide a rubric for characteristics that distinguish the end of National Banking Era panics.
Donaldson’s findings parallel recent work by Gorton (2008, 2009, 2010) on the recent financial crisis showing how financial markets behaved differently once the panic conditions ensue. Gorton (2009, 2010) argues that the conventional suspect—sub-prime mortgage-backed assets—can explain the spark of the crisis but that the financial crisis widened across asset classes as counter-party risk spread among financial market participants.
Unable to locate data on inter-regional cash flows from NYC during earlier panics, we rely on anecdotal references for earlier panics. Sprague (1910: 173) writes that by mid-July 1893 New York City banks lost $40 million in interior deposits and $20 million in reserves (referring to the May 4 and the July 12 call reports), which he attributes to western and southern banks withdrew their cash. Wicker (2000: 29) writes that during the panic of 1873, interior banks reduced their deposits at reserve agents by $42 million, and New York City banks released $14.5 million of legal tender to the interior. Relative to a two year moving average for reserves held at New York Clearing House banks, $14.5 million in cash represented nearly 22% in 1873 and $20 million was about 15% of average reserves of New York Clearing House banks in 1893. The accumulated cash outflow from NYCH banks in 1907 was more than $100 million and over 30% of pre-panic reserve balances.
Wicker (2000) provides an extensive investigation into the key events that preceded the Panic of 1907.
These data are from the New York Tribune and we thank Caroline Fohlin for assistance in getting them.
Annual percentage rate.
Gorton and Tallman (2016) are unable to document any instance in which there is documentation verifying a decision to suspend convertibility among the New York Clearing House banks.
We have estimates of the premium on currency in New York City as collected by Andrew (1908). We use these series in graphics and lower frequency estimation because the series has missing observations on a daily basis that would reduce notably the number of observations for our empirical investigation using daily data. Of 47 potential observations, there are 12 missing, which increases if the series are lagged.
Schwert (1990) uses the components of the Dow Jones Index of industrial and transportation (mainly railroads) stock equity prices for which the prices (not the number of shares outstanding) are the inputs for the weights. The stocks in the index are among the most heavily traded and represent “high-quality” stock investments. Over a more limited sample [September 30, 1907, to February 22, 1908], we have data from Fohlin et al. (2016), who compute an index as a simple average of all “last” prices for every common stock on every trading day during the period. The prices are dollar prices and are not adjusted for dividends or stock splits. The latter provide a wider range of stock prices in its computation. Econometric results employ the Schwert measure, although results are robust to using either measure.
We refer to banking aggregates and daily transactions between banks. The ledger books of the New York Clearing House are available at the Columbia University Rare Manuscripts Library. They provide information regarding end of day net balances positions between member banks and the New York Clearing House and may be helpful for gauging the cash drains from individual banks.
We have results that extend the endpoint January 11, 1908 (62 observations). The results are less consistent, but the call rate fluctuations are relatively modest over the extended period.
We note that reducing the sample endpoint to Saturday, December 7, 1907, reduces the estimation period to one in which there is the most volatility in the call money interest rate. Over that sample, the regressions indicate a more influential role for clearing house loan certificates on the call money interest rate. Hoag (2015) notes that clearing house loan certificates requested (our data) may not have been the volume that banks circulated within the New York Clearing House, and may be a source of measurement error. Hence, coefficient estimates may be subject to attenuation bias. We thank an anonymous referee for the reference and point.
We tested the call loan rate for a unit root using an extended time series (January 2, 1907, to December 31, 1908) in an application of the Augmented Dickey–Fuller test. We reject the null hypothesis for I(1) at standard levels of confidence. We thank an anonymous referee for the suggestion.
New York Times, November 6, 1893. Exhibit follows Appendix 1B.
Call report data highlight only the remarkable recovery of reserves and net deposits among New York Clearing House Banks in the February 1908 report relative to the lows observed in the December 3, 1907, report.
We use clearing house loan certificates as change in levels to be consistent with the net gold flow variable, which is only measured as a change in level.
Call rate observations are calculated as an average over the week and dated as compiled on Fridays. Gold data were compiled on Saturdays. Given that net gold inflows and clearing house loan certificates are proposed to be related, we choose Saturday observations for the clearing house loan certificates data. All right-hand side variables are lagged. These timing issues are noted but not addressed. We use the weekly data to calculate the reserve requirements, which may differ from the data used by the Comptroller of the Currency (for national banks) and the New York State Superintendent of Banking (for state banks) in their official calculations, hence our qualification for “estimated” required reserves.
All New York Clearing House member banks faced a 25% reserve requirement, which is the rate faced by New York City national banks (central reserve city banks).
The lack of significance of clearing house loan certificates in the call loan regression does not nullify their importance during the panic; their issuance facilitated the imports of gold to the USA, and the daily data results are consistent with a palliative role in providing credit to the local financial system. Furthermore, the response of the call loan rate to changes in net issues of clearing house loan certificates likely was not symmetric. That is, the withdrawal of loan certificates did not result in an increase in the call loan rate. Loan certificates were being withdrawn because they were no longer needed as liquidity returned to banking. We thank an anonymous referee for highlighting this point.
Recall that Donaldson used the stock market index return along with the reserve, deposits and loan data to explain the call loan interest rate over an extended sample period. By focusing on the 1907 period, we add the net gold import data and the clearing house loan certificates data to the analysis.
In the example, $4.86656–$4.83 = 0.03656, a 0.7515% return. The return exceeded the gold shipping costs, so it could generate gold shipments. The return seemed small, yet it was a lucrative activity for large market participants. If the activity took 2–3 weeks to complete and could be replicated repeatedly, the annual return on 0.75% would be between 14 and 22% on an annual basis. A currency premium of as much as 4% made this transaction all the more attractive.
Silber (2008, p. 55).
The contemporary analysis appearing in the Commercial and Financial Chronicle and the New York Times emphasizes that the much of the gold imported into New York City flowed directly to the interior. There are good reasons to believe that these are accurate reports. Firstly, the typical flow of funds in the fall was toward agricultural producers in the interior of the country from New York City, the intermediary between the payments from European importers of agricultural products and the US interior agricultural producers. Secondly, the US trade surplus in 1907 hovered somewhere near $200 million, so profitable importation of gold to the USA was consistent with aggregate trade flows.
This observation is a stark contrast with what was observed in 2008 in the aftermath of the failure of Lehman Brothers. After September 15, 2008, following the failure of Lehman Brothers, the daily decline of over 4% (for the Dow Jones and S&P 500 indexes) was followed by a subsequent net decline of over 35% by March 16, 2009.
The standard deviation of weekly stock returns for the constructed index was approximately 2.0%. Although the nadir of the Fohlin series is reached on November 21 versus November 22, we find the same “relative” differences, namely, a 4% difference between the nadir of the series and the level on October 24th) using the Fohlin stock index measure. It is notable that despite a much higher volatility in the Fohlin series relative to the Schwert (Dow Jones) series, the two series have comparable indicator properties.
The Long Island unified gold fours of 1949 goes into receivership in 1908, and for consistency, we leave this series out of the bond index. It could easily be accommodated, but its absence likely has little effect on our results.
We look at the capital gain return only, abstracting from dividend yields, because we are only interested in detecting large negative returns.
See Bruner and Carr (2007).
See analysis below for verification.
For example, the observation of April 27, 1906, largely reflects the ramifications from the San Francisco earthquake of April 19, 1906 (see Odell and Weidenmier 2004).
Gorton and Tallman (2016) find co-movement between the stock and bond volatility measures for the Panic of 1893.
Gorton and Tallman (2016) investigate observable financial market characteristics that determine when deposits from interior bankers return to the New York City national banks.
References
Andrew AP (1908) Hoarding in the Panic of 1907. Q J Econ 22(2):290–299
Andrew AP (1910) Statistics for the United States. Government Printing Office, Washington
Bruner RF, Carr S (2007) The Panic of 1907: Lessons learned from the market’s perfect storm. Wiley, New York
Calomiris CW, Gorton G (1991) The origins of banking panics: models, facts, and bank regulation. In: Glenn Hubbard R (ed) Financial markets and financial crises. University of Chicago Press, Chicago, pp 109–173
Carlson M (2005) Causes of bank suspensions in the panic of 1893. Explor Econ Hist 42(1):56–80
Carosso VP (1987) The Morgans: Private International Bankers, 1854–1913. Harvard University Press, Cambridge
Champ BA (2007) The national banking system: a brief history. Federal Reserve Bank of Cleveland Working Paper No. 07-23. Available at SSRN: http://ssrn.com/abstract=1171222
Coleman A (2007) The pitfalls of estimating transactions costs from price data: evidence from trans-Atlantic gold-point arbitrage, 1886–1905. Explor Econ Hist 44:387–410
Donaldson RG (1992) Sources of panics: evidence from the weekly data. J Monet Econ 30:277–305
Dupont B (2009) Panic in the plains: agricultural markets and the panic of 1893. Cliometrica 3(1):27–54
Fohlin C, Gehrig T, Haas M (2016) Rumors and Runs in Opaque Markets: Evidence from the Panic of 1907. Emory economics 1605, Department of Economics, Emory University (Atlanta)
Friedman M, Schwartz AJ (1963) A monetary history of the United States, 1867–1960. Princeton University Press for NBER, Princeton
Goodhart CAE (1969) The New York Money Market and the Finance of Trade, 1900–1913. Harvard University Press, Cambridge
Gorton GB (1988) Bank panics and business cycles. Oxford Economic Papers, New Series 40:751–781
Gorton GB (2008) The panic of 2007. In: Proceedings of Conference entitled Maintaining financial stability in a changing financial system, Federal Reserve Bank, Kansas, pp 131–262
Gorton GB (2009) Slapped in the face by the invisible hand: banking and the panic of 2007. Yale University Working Paper. Available at SSRN: http://ssrn.com/abstract=1401882
Gorton GB (2010) Slapped by the invisible hand: the panic of 2007. Oxford University Press, New York
Gorton GB, Tallman, EW (2016) How did pre-fed banking panics end? NBER Working Paper 22036
Hoag C (2005) Deposit drains on “interest-paying” banks before financial crises. Explor Econ Hist 42(4):567–585
Hoag C (2015) Clearinghouse loan certificates as interbank loans. Trinity College Department of Economics Working Paper No. 1054
Huebner S (1922) The stock market. D. Appleton, New York
James JA, McAndrews J, Weiman DF (2013) Wall street and main street: the macroeconomic consequences of new york bank suspensions, 1866 to 1914. Cliometrica 7(2):99–130
Kemmerer ER (1910) Seasonal variations in the relative demand for money and capital in the United States. Government Printing Office, Washington
Kindleberger C (2000) Manias, panics, and crashes: a history of financial crises. Wiley, New York
Laughlin JL (1912) Banking reform. National Citizen’s League
Macaulay F (1938) Some theoretical problems suggested by the movements of interest rates, bond yields and stock prices in the United States since 1856. National Bureau of Economic Research, Cambridge
Meeker JE (1922) The work of the stock exchange. Ronald Press Company, New York
Miron J (1986) Financial panics, the seasonality of the nominal interest rate, and the founding of the fed. Am Econ Rev 76:125–140
Muhleman ML (1908) Monetary and banking systems. Monetary Publishing Company, New York
New York Sun, New York Times, New York Tribune, various issues
Noyes, AD (1909) Forty years of american finance. G.P Putnam and Sons, New York
Odell K, Weidenmier M (2004) Real shock, monetary aftershock: the 1906 San Francisco earthquake and the panic of 1907. J Econ Hist 64(4):1002–1027
Officer L (2008) Gold Standard. EH.Net Encyclopedia, edited by Robert Whaples. http://eh.net/encyclopedia/article/officer.gold.standard. 26 March 2008
Pratt S (1903) The work of wall street. D. Appleton and Company, New York
Rich G (1989) Canadian banks, gold, and the crisis of 1907. Explor Econ Hist 26(2):135–160
Rodgers MT, Wilson BK (2011) Systemic risk, missing gold flows, and the panic of 1907. Q J Aust Econ 14(2):158–187
Schwert GW (1990) Indexes of United States stock prices from 1802 to 1987. J Bus 63:399–426
Silber WL (2008) When washington shut down wall street: the great financial crisis of 1914 and the origins of America’s Monetary Supremacy. Princeton University Press, Princeton
Sprague OMW (1910) History of crises under the national banking system. Government Printing Office, Washington
Wicker ER (2000) The banking panics of the guilded age. Cambridge University Press, New York
Acknowledgements
The authors thank the participants of the conference and the session, especially Jerry Dwyer, Mary O’Sullivan, and Giovanni Toniolo. We also thank Ben Craig and Margaret Jacobson for helpful conversations. We also thank two anonymous referees for valuable suggestions.
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix 1A: Data sources
1.1 Bond prices
Weekly: Kemmerer (1910), Appendix O, Tables 21–47, pp. 413–508 (see Appendix 1B).
1.2 Call loan interest rates
Daily observations accumulated from the New York Tribune Financial Page, Money and Exchange column, from August 31, 1907, through February 21, 1908. Source provides the range of interest rates observed over the day, describe the “ruling rate”—the rate at which the bulk of trades were made. More complete source than data we accumulated from the New York Times.
Weekly observations are taken from Kemmerer (1910), Appendix A, Table 1, pp. 235–236.
Monthly observations are from Macaulay (1938), Table 10, page A141.
1.3 Clearing house loan certificates
Daily observations are taken from the minutes of the meetings of the Clearing House loan committee of the New York Clearing House. These are indicated by date, by issuer (requesting bank), and by amount. Data begin October 26, 1907, through April 4, 1907. Significant transactions end mid-January 1908.
1.4 Commercial paper interest rate
Monthly observations are from Macaulay (1938), Table 10, page A141.
1.5 Currency premium
Measured daily, by Andrew (1908).
1.6 Exchange rates
Weekly observations are from Andrew (1910), Table 12, p. 177.
1.7 Panic of 1907 dummy variable
Daily: October 22, 1907, until January 11, 1908, = 0, otherwise.
Weekly: = 1 from October 26, 1907, until January 11, 1908, = 0 at all other dates.
Monthly: = 1 for October, November, December 1907, January 1908; = 0 otherwise.
Applies to regressions and to graphics.
1.8 Reserves of New York Clearing House Banks
Taken from Andrew (1910), Table 31, Weekly Statement of New York Clearing House Banks, pp. 75–118.
1.9 Stock returns
We use the daily returns data from Schwert (1990) for the daily and weekly analysis.
Monthly observations are from the Cowles Commission Common Stock Index taken from the NBER website.
1.10 Gold
Weekly: Kemmerer (1910), Appendix L, Imports and Exports of Gold, Table 18, p. 389.
Monthly: Kemmerer (1910), Appendix L, Imports and Exports of Gold, Table 17, p. 387.
Appendix 1B: Specific bond series descriptions
- Bond 1 :
-
Atchison, Topeka and Santa Fe adjustment gold fours of 1995:1896–1908. Coupon payments: May 1, November 1; bond 109 in Macaulay (1938). A bond included in Rodgers and Wilson (2011)
- Bond 2 :
-
Atchison, Topeka and Santa Fe general gold fours of 1995:1897–1908. Coupon payments: April 1, October 1; bond 94 in Macaulay (1938)
- Bond 3 :
-
Baltimore and Ohio gold fours of 1948: 1900–1908. Coupon payments: April 1, October 1; Bond 90 in Macaulay (1938)
- Bond 4 :
-
Central Pacific first refunding gold fours of 1949: 1900–1908. Coupon payments: February 1, August 1; bond 101 in Macaulay (1938). A bond included in Rodgers and Wilson (2011)
- Bond 5 :
-
Central Railroad of New Jersey general gold fives of 1987: 1890–1908. Coupon payments: January 1, July 1; bond 104 in Macaulay (1938)
- Bond 6 :
-
Chesapeake and Ohio general gold four-and-a-half of 1992: 1893–1908. Coupon payments: March 1, September 1; bond 135 in Macaulay (1938). A bond included in Rodgers and Wilson (2011)
- Bond 7 :
-
Chicago, Burlington and Quincy (Nebraska extension) fours of 1927: 1890–1908
- Bond 8 :
-
Chicago, Milwaukee and St. Paul general gold fours of 1989: 1890–1908. Coupon payments: January 1, July 1; bond 87 in Macaulay (1938)
- Bond 9 :
-
Denver and Rio Grande first consolidated gold fours of 1936: 1890–1908
- Bond 10 :
-
Erie first consolidated gold fours prior lien of 1996: 1898–1908. Coupon payments: January 1, July 1; bond 146 in Macaulay (1938)
- Bond 11 :
-
Hocking Valley first consolidated gold four-and-a-half of 1999: 1900–1908. Coupon payments: January 1, July 1; bond 105 in Macaulay (1938)
- Bond 12 :
-
Iowa Central first gold fives of 1938: 1890–1908
- Bond 13 :
-
Long Island unified gold fours of 1949: 1900–1908
- Bond 14 :
-
Louisville and Nashville unified gold fours of 1940: 1898–1908
- Bond 15 :
-
Missouri, Kansas and Texas first gold fours of 1990: 1891–1908. Coupon payments: June 1, December 1; bond 134 in Macaulay (1938)
- Bond 16 :
-
Missouri Pacific first consolidated gold sixes of 1920: 1890–1908
- Bond 17 :
-
Missouri Pacific, St. Louis, Iron Mountain and Southern general consolidated gold fives of 1931: 1894–1908
- Bond 18 :
-
New York Central and Hudson River gold three-and-a-half of 1997: 1899–1908. Coupon payments: January 1, July 1; bond 95 in Macaulay (1938)
- Bond 19 :
-
New York Central and Hudson River (West Shore) first fours guaranteed of 2361: 1890–1908. Coupon payments: January 1, July 1; bond 70 in Macaulay (1938)
- Bond 20 :
-
New York, Ontario and Western refunding first gold fours of 1992: 1893–1908
- Bond 21 :
-
Norfolk and Western first consolidated gold fours of 1996: 1897–1908. Coupon payments: April1, October 1; bond 103 in Macaulay (1938)
- Bond 22 :
-
Northern Pacific prior lien gold fours of 1997: 1897–1908. Coupon payments: January 1, April 1, July 1, October 1; bond 102 in Macaulay (1938). Included in Rodgers and Wilson (2011)
- Bond 23 :
-
St. Louis and San Francisco general gold fives of 1931: 1890–1908
- Bond 24 :
-
St. Louis and Southwestern first gold fours of 1989: 1892–1908. Coupon payments: May 1, November 1; bond 133 in Macaulay (1938)
- Bond 25 :
-
Southern Railway first consolidated fives of 1994: 1895–1908. Coupon payments: January 1, July 1; bond 119 in Macaulay (1938)
- Bond 26 :
-
Union Pacific land grant gold fours of 1947: 1899–1908. Coupon payments: January 1, July 1; bond 91 in Macaulay (1938)
- Bond 27 :
-
Wabash first gold fives of 1939: 1890–1908. Coupon payments: May 1, November 1; bond 78 in Macaulay (1938)
Note All bond data series are taken from Kemmerer (1910). The price series for these selected bonds were chosen on the suggestion of investment bankers as most likely to have traded in liquid markets. Kemmerer notes (page 174) that quotations for prices were one of the following in order of preference: (1) Prices at which sales took place (on Friday or closest day), (2) mean of “bid” and “asked” prices on Friday (or closest day), (3) “bid” quotations on Friday, (4) “Asked” quotations on Friday, and (5) if no quotations, the mean of the prices at nearest week.
Appendix 2: Time series properties of data series
Daily Data: October 26, 1907, until January 4, 1908 Call loan rate: Sample mean: 13.7, standard error: 9.98 Autocorrelations | |||||||||||||
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 |
0.82 | 0.75 | 0.69 | 0.63 | 0.56 | 0.59 | 0.47 | 0.37 | 0.30 | 0.26 | 0.18 | 0.18 | −0.14 | 0.07 |
Differenced call loan rate: Sample mean: −0.01, standard error: 0.068 Autocorrelations | |||||||||||||
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 |
−0.22 | −0.25 | 0.24 | 0.02 | −0.31 | 0.32 | 0.03 | −0.37 | −0.01 | 0.15 | −0.18 | −0.04 | 0.24 | −0.23 |
Log of clearing house loan certificates: Sample mean: 4.2, standard error: 0.54 Autocorrelations | |||||||||||||
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 |
0.98 | 0.95 | 0.89 | 0.82 | 0.56 | 0.74 | 0.57 | 0.48 | 0.39 | 0.31 | 0.24 | 0.17 | 0.11 | 0.05 |
Percent change in clearing house loan certificates: Sample mean: −0.01 standard error: 0.068 Autocorrelations | |||||||||||||
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 |
0.75 | 0.54 | 0.33 | 0.22 | 0.23 | 0.17 | 0.14 | 0.08 | −0.01 | 0.04 | 0.04 | 0.06 | 0.04 | −0.04 |
Log of stock index: Sample mean: 4.3, standard error: 0.036 Autocorrelations | |||||||||||||
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 |
0.93 | 0.87 | 0.80 | 0.73 | 0.65 | 0.55 | 0.44 | 0.35 | 0.28 | 0.23 | 0.20 | 0.15 | 0.12 | 0.12 |
Stock return: Sample mean: 0.05 standard error: 1.4 Autocorrelations | |||||||||||||
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 |
−0.04 | 0.2 | −0.07 | 0.07 | 0.19 | 0.12 | −0.12 | −0.04 | −0.20 | −0.05 | 0.00 | −0.05 | −0.08 | −0.05 |
Short daily sample: October 28, 1907, until November 30, 1907 (29 observations) Call loan rate: Sample mean: 13.9, standard error: 10.05 Autocorrelations | |||||||||||||
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 |
0.84 | 0.71 | 0.60 | 0.47 | 0.32 | 0.28 | 0.21 | 0.06 | −0.04 | −0.14 | −0.18 | −0.20 | −0.19 | −0.18 |
Differenced call loan rate: Sample mean: −1.04 standard error: 5.28 Autocorrelations | |||||||||||||
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 |
−0.19 | −0.14 | 0.07 | 0.15 | −0.28 | 0.08 | 0.38 | −0.12 | −0.00 | −0.15 | 0.01 | −0.13 | −0.05 | 0.07 |
Log of clearing house loan certificates: Sample mean: 4.24, standard error: 0.31 Autocorrelations | |||||||||||||
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 |
0.95 | 0.82 | 0.67 | 0.52 | 0.38 | 0.26 | 0.16 | 0.08 | 0.01 | −0.04 | −0.12 | −0.17 | −0.18 | −0.25 |
Percent change in clearing house loan certificates: Sample mean: −0.057 standard error: 0.099 Autocorrelations | |||||||||||||
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 |
0.85 | 0.71 | 0.59 | 0.42 | 0.31 | 0.22 | 0.09 | −0.01 | −0.08 | −0.10 | −0.12 | −0.16 | −0.18 | −0.26 |
Log of stock index: Sample mean: 4.3, standard error: 0.023 Autocorrelations | |||||||||||||
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 |
0.75 | 0.59 | 0.36 | 0.18 | 0.01 | −0.21 | −0.37 | −0.51 | −0.54 | −0.48 | −0.38 | −0.43 | −0.36 | −0.28 |
Stock return: Sample mean: 0.001 standard error: 1.5 Autocorrelations | |||||||||||||
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 |
−0.15 | 0.19 | −0.01 | 0.07 | 0.25 | −0.09 | 0.08 | −0.15 | −0.16 | −0.08 | 0.11 | −0.22 | −0.09 | −0.20 |
Weekly data: January 6, 1899, until December 26, 1908 Net imports of gold: Sample mean = − 0.270, standard error = 2.9 Autocorrelations | |||||||||||
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 |
0.69 | 0.52 | 0.44 | 0.28 | 0.19 | 0.12 | 0.11 | 0.10 | 0.05 | 0.02 | 0.04 | 0.02 |
Weekly data: January 6, 1900, until December 26, 1908 Call loan rate: Sample mean = 4.126, standard error = 4.43 Autocorrelations | |||||||||||
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 |
0.58 | 0.46 | 0.28 | 0.22 | 0.18 | 0.24 | 0.22 | 0.21 | 0.14 | 0.07 | 0.05 | 0.06 |
Weekly data: January 7, 1899, until December 26, 1908 Reserve surplus or deficit: Sample mean = 15.418, standard error = 15.99 Autocorrelations | |||||||||||
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 |
0.95 | 0.87 | 0.80 | 0.72 | 0.65 | 0.58 | 0.52 | 0.48 | 0.44 | 0.40 | 0.36 | 0.35 |
Weekly data: January 13, 1899, until December 26, 1908 Difference in reserve surplus or deficit: Sample mean = − 0.13 Autocorrelations | |||||||||||
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 |
0.34 | −0.04 | −0.08 | 0.05 | −0.02 | −0.16 | −0.15 | −0.02 | 0.07 | −0.14 | −0.19 | −0.07 |
Weekly data: January 7, 1899, until December 26, 1908 Stock market index—returns without dividends: Sample mean = 0.001, standard error = 0.022 Autocorrelations | |||||||||||
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 |
0.04 | 0.04 | −0.03 | 0.00 | −0.02 | −0.04 | 0.004 | −0.04 | 0.16 | −0.00 | 0.04 | 0.09 |
Weekly data: January 6, 1900, until December 26, 1908 Holding period returns to the bond index: Sample mean: 0.000, standard error = 0.0032 Autocorrelations | |||||||||||
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 |
0.34 | 0.04 | 0.13 | 0.11 | −0.10 | −0.14 | −0.02 | 0.01 | 0.16 | −0.01 | 0.04 | 0.01 |
Correlations for daily data, Sample: October 26, 1907, to November 30, 1907 (results for the sample ending January 4, 1908, are not materially different)
Covariance/correlation matrix | Call rate | Log CHLC | Log of stock index |
|---|---|---|---|
Call rate | 0.033 | 0.75 | 0.806 |
Log CHLC | 0.573 | 17.77 | 0.99 |
Log of stock index | 0.63 | 18.08 | 18.56 |
Covariance/correlation matrix | Differenced call rate | Stock Return | Percent change in clearing house loan certificates |
|---|---|---|---|
Differenced call rate | 2.662e−003 | 0.041 | −0.143 |
Stock return | 3.107e−005 | 2.156e−004 | −0.098 |
Percent change in clearing house loan certificates | −9.558e−004 | −1.869e−004 | 1.674e−002 |
Correlations for weekly data, Sample: January 6, 1900, to December 26, 1908
Covariance/correlation matrix | Call interest rate | Change in the Reserve Surplus or Deficit | Net Gold Inflows (millions) |
|---|---|---|---|
Call interest rate | 39.2278126 | −0.14328 | 0.08499 |
Change in the Reserve Surplus or Deficit | −4.5090284 | 25.2479431 | 0.09695 |
Net Gold Inflows (millions) | 1.6119909 | 1.4751372 | 9.1701615 |
Correlations for weekly data, Sample: October 26, 1907, to January 4, 1908
Covariance/correlation matrix | Call interest rate | Change in the reserve surplus or deficit | Net gold inflows (millions) | Change in Clearing house loan certificates |
|---|---|---|---|---|
Call interest rate | 524.82 | −0.62 | 0.36 | 0.86 |
Change in the reserve surplus or deficit | −195.06 | 191.61 | 0.13 | −0.91 |
Net gold inflows (millions) | 88.25 | 18.97 | 116.93 | 0.17 |
Change in clearing house loan certificates | 283.12 | −180.76 | 26.84 | 206.73 |
Appendix 3: Bond and stock return co-movements
Over the entire sample period, there were 13 instances of negative co-movements and 7 instances of positive co-movements. There three instances of consecutive weekly observations. One in the negative direction (bold) and one in the positive direction (italics) occurred during the critical weeks in the Panic of 1907. One in the positive direction (italics) occurred in November 1900 following the reelection of McKinley. Dates of the instances are listed below:
May 12, 1900 (negative)
June 23, 1900 (negative)
October 20, 1900 (positive)
November 10, 1900 (positive)
November 17, 1900 (positive)
May 11, 1901 (negative)
July 13, 1901 (negative)
September 19, 1903 (negative)
April 28, 1906 (negative)
March 9, 1907 (negative)
April 6, 1907 (positive)
August 17, 1907 (negative)
October 19, 1907 (negative)
October 26, 1907 (negative)
November 16, 1907 (negative)
November 30, 1907 (positive)
December 7, 1907 (positive)
February 8, 1908 (negative)
May 16, 1908 (positive)
May 23, 1908 (negative)
Rights and permissions
About this article
Cite this article
Tallman, E.W., Moen, J.R. The transmission of the financial crisis in 1907: an empirical investigation. Cliometrica 12, 277–312 (2018). https://doi.org/10.1007/s11698-017-0161-1
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11698-017-0161-1