I’ve completed a new historical investigation of early stock market returns, now posted on SSRN: https://papers.ssrn.com/sol3/papers.cfm ... id=5762842
Aside from those curious about older market history, the paper may be of interest to forum members interested in the pricing of risk and the contribution of dividends.
Here is the opening, to give the flavor:
"Ground-breaking work by Richard Sylla and colleagues has greatly expanded the supply of information on early securities markets in the US (Sylla, Wilson and Wright 2005). A data record that formerly petered out about 1871 (Cowles 1939), and was, even recently, scanty back to 1825, and confined to New York only (Goetzmann, Ibbotson and Peng 2001), now extends back to the early 1790s across Boston and Philadelphia as well as New York.
Unfortunately for studies of investment return, the Sylla database contains only a bare record of price action transcribed from newspapers. It contains no information on share count, dividend, or firm fate. Hilt (2009a, 2009b) was among the first to leverage the Sylla record to identify firms prominent enough to repay further investigation. In that era there were hundreds of banks, and many dozen insurance companies, that were too small and obscure to have left a trading record or even have an exchange listing (Davis 1917; Sylla and Wright 2013). With the firms that did trade on the New York exchange identified from the Sylla files, Hilt tapped archival resources to identify both the firms that failed and the date of failure. This supplemental information allowed Hilt to develop the extent and severity of the hitherto obscure Panic of 1826, in which a dozen insurance firms in New York failed. Although Hilt was the first to report share counts for New York stocks as of 1826, he did not report dividends and did not attempt to construct an index of investor return.
McQuarrie (2024a), working almost twenty years after the Goetzmann and Sylla efforts, benefitted from the subsequent digitization of old newspaper archives in the early and mid-2000s. Using diverse sources, including Hilt’s tabulation, he was able to combine the Sylla information on price action with information on the share count of each firm and also to compile a dividend record for each traded stock. He produced the first value-weighted index of the total return on US stocks prior to the Civil War, drawing on the Sylla price record for Baltimore, Boston and Philadelphia as well as New York. Banks and transportation stocks dominated the index.
In an early phase of that work McQuarrie (2018) had compiled share count and dividend information on the insurance sector as well. However, he later chose not to include insurance stocks in his reported stock market index. The Great Fire of December 16th, 1835 in New York posed a conundrum. Most of the two dozen fire insurance firms in New York failed that day (Geldert 1906). If the 100% decline for these stocks were to be blended into the total stock market index on a value-weighted basis, returns on that index would be depressed further. McQuarrie’s market returns were already substantially below those reported in Siegel (2022), based on later failures omitted from Siegel’s data, such as that of the 2nd Bank of the United States during the Panic of 1837. He made a strategic decision to set aside the insurance stocks, to protect the credibility of the overall effort.
This paper revisits that decision. I compile a monthly index of total return on insurance stocks in New York from January 1799 to January 1845, with failures identified and their decline in price incorporated into the data. I proceed to calibrate the extent to which adding the insurance sector on a value-weighted basis would further depress the total market index return for that period as reported in McQuarrie (2024a). To start, I describe the early decades of trading in New York, and situate insurance stocks relative to the other sectors traded."
Aside from those curious about older market history, the paper may be of interest to forum members interested in the pricing of risk and the contribution of dividends.
Here is the opening, to give the flavor:
"Ground-breaking work by Richard Sylla and colleagues has greatly expanded the supply of information on early securities markets in the US (Sylla, Wilson and Wright 2005). A data record that formerly petered out about 1871 (Cowles 1939), and was, even recently, scanty back to 1825, and confined to New York only (Goetzmann, Ibbotson and Peng 2001), now extends back to the early 1790s across Boston and Philadelphia as well as New York.
Unfortunately for studies of investment return, the Sylla database contains only a bare record of price action transcribed from newspapers. It contains no information on share count, dividend, or firm fate. Hilt (2009a, 2009b) was among the first to leverage the Sylla record to identify firms prominent enough to repay further investigation. In that era there were hundreds of banks, and many dozen insurance companies, that were too small and obscure to have left a trading record or even have an exchange listing (Davis 1917; Sylla and Wright 2013). With the firms that did trade on the New York exchange identified from the Sylla files, Hilt tapped archival resources to identify both the firms that failed and the date of failure. This supplemental information allowed Hilt to develop the extent and severity of the hitherto obscure Panic of 1826, in which a dozen insurance firms in New York failed. Although Hilt was the first to report share counts for New York stocks as of 1826, he did not report dividends and did not attempt to construct an index of investor return.
McQuarrie (2024a), working almost twenty years after the Goetzmann and Sylla efforts, benefitted from the subsequent digitization of old newspaper archives in the early and mid-2000s. Using diverse sources, including Hilt’s tabulation, he was able to combine the Sylla information on price action with information on the share count of each firm and also to compile a dividend record for each traded stock. He produced the first value-weighted index of the total return on US stocks prior to the Civil War, drawing on the Sylla price record for Baltimore, Boston and Philadelphia as well as New York. Banks and transportation stocks dominated the index.
In an early phase of that work McQuarrie (2018) had compiled share count and dividend information on the insurance sector as well. However, he later chose not to include insurance stocks in his reported stock market index. The Great Fire of December 16th, 1835 in New York posed a conundrum. Most of the two dozen fire insurance firms in New York failed that day (Geldert 1906). If the 100% decline for these stocks were to be blended into the total stock market index on a value-weighted basis, returns on that index would be depressed further. McQuarrie’s market returns were already substantially below those reported in Siegel (2022), based on later failures omitted from Siegel’s data, such as that of the 2nd Bank of the United States during the Panic of 1837. He made a strategic decision to set aside the insurance stocks, to protect the credibility of the overall effort.
This paper revisits that decision. I compile a monthly index of total return on insurance stocks in New York from January 1799 to January 1845, with failures identified and their decline in price incorporated into the data. I proceed to calibrate the extent to which adding the insurance sector on a value-weighted basis would further depress the total market index return for that period as reported in McQuarrie (2024a). To start, I describe the early decades of trading in New York, and situate insurance stocks relative to the other sectors traded."
Statistics: Posted by McQ — Thu Nov 20, 2025 11:47 pm — Replies 0 — Views 85