The Shanghai Bund and the Huangpu River photographed in the 1860s or 1870s: the settlement skyline along the water, with steamers, lighters and cargo junks moored off the wharves.
The Bund and the Huangpu. Photograph by Lai Afong (黎芳), c. 1860–80. China Merchants’ Steamship Navigation Company — the largest stock in the tables below — bought its head office at No. 9 the Bund, together with the wharves and the fleet, in 1877. The Chinese-owned share market itself traded inland, in the walled city. Public domain, via Wikimedia Commons.

申報 Shen Bao · 平準公司各股份市價 · 1882–83

A daily record of Qing China’s domestic stock market crash

Shanghai’s Chinese-owned share market boomed through 1882 and broke in 1883, taking the native banks with it. This is every price the market printed across that rise and fall — all 33 listed companies, one trading day at a time, read off the scanned page. The existing record for this market holds one price per company per month.

How the index is built

Three constructions of the market from the same daily panel: a repeat-sales index, a fixed basket, and a chained average. They are shown together because a single construction cannot tell you whether its own answer is an artefact.

Repeat-sales regresses every pair of quotes for the same security on date dummies — 35,383 equations — so the index is anchored to observed price levels rather than compounded from day to day. Fixed basket averages each firm’s price against its own base over constant membership. Chained compounds the daily cross-sectional mean return. The three are shown together because the choice is not cosmetic: as this panel was corrected through five rounds of re-extraction, the repeat-sales answer stayed inside a three-point band while the chained one moved across twenty-three, which is what it means for an estimator to carry every mis-read cell forward permanently. Dates priced by fewer than four securities are interpolated rather than fitted.

Where the money was, and where it went

The same panel, split by what the companies actually did. Each line is a repeat-sales index over the securities in that industry, fitted exactly as the market index is, so the sector lines and the market line are comparable by construction.

Counts are the companies each index is actually fitted on, not every company in the sector: a listing quoted too rarely to pair against itself carries no information for a repeat-sales fit and is left out, which is why manufacturing shows six of its eight names. Land is dropped entirely — one company.

Daily prices

Click a company to show or hide it. Prices as printed — taels for most, yuan or pounds for a few — so the vertical scale is a level, not an index.

One point per issue carrying the table, in the order the paper printed them. Isolated mis-read cells removed; a price is dropped only when it disagrees with the quotes either side of it and those two agree with each other, so genuine step changes survive.

Daily quotes

Select a single company above to list every price it was quoted at.

Coverage by security

Trading days captured, first to last quote, and the fall from each company’s own peak.

There is not one bank on this list. Thirty-three companies are quoted here — mines, mills, steamers, insurers, a telegraph, a waterworks — and no financial institution among them. That is not an omission in the newspaper. Shanghai’s native banks, the qianzhuang (錢莊), were sole proprietorships and partnerships carrying unlimited liability. They had no transferable shares to quote. China would not have a joint-stock, limited-liability bank until the Imperial Bank of China opened on 27 May 1897, fourteen years after these tables were printed.

Which is exactly the sector that broke. The crash of 1883 ran through the credit system this list cannot show. When Hu Xueyan’s corner in raw silk collapsed that year, he owed some 560,000 taels to around forty Shanghai qianzhuang, and most of them failed. The prices below are the borrowers; the lenders are invisible. A reader watching only this market would have seen the collateral fall without ever seeing the leverage that was forcing it.

The insurers are the closest thing to a financial sector here, being the only listed businesses holding portfolios of financial claims rather than mines and machinery. They behave unlike the rest: insurance peaks latest of any industry and finishes the window roughly where it started, while every other sector ends below par. Whether that reflects genuine resilience or simply thinner, stickier quotation is not something these prices can settle.

The boom was a mining flotation. Fourteen of the thirty-three companies are mines — copper, coal, iron, silver — against three steamship and dock companies and three utilities. Sector indexes make the rotation visible: the peaks are staggered rather than simultaneous, and mining, the crowd’s favourite, is not the first to turn.

A postscript. When China’s first joint-stock bank was finally floated in 1897, two of its major shareholders were the China Merchants’ Steamship Navigation Company and the Telegraph Administration — both of them stocks in this table. The market that could not list a bank in 1882 supplied the capital for one fifteen years later.