#285 1823 · Rishengchang (日升昌) · Banking and remittance
Rishengchang stopped moving silver across bandit country and moved a coded slip of paper instead
the problem
Merchants had no safe way to move silver between distant cities
background
In the early nineteenth century, a Chinese merchant with money in one city and business in another had one real option: send the silver itself, under armed escort, along trade routes thick with bandits. Guards, insurance and the occasional total loss were a standing cost of doing business at any distance, and the risk did not fall as volume grew — every shipment was its own gamble.
Lei Lutai, manager of the Pingyao dye firm Xiyucheng, proposed converting it into a dedicated remittance house instead: a merchant would deposit silver at one branch and draw the equivalent at another on a paper draft, so the thing that traveled was ink, not metal. To stop the drafts themselves from being forged or cashed by whoever intercepted them, the sum, date and unit were encoded with rotating poem-based ciphers in place of plain numerals — a code the receiving branch could check and a thief holding the paper could not.
what everyone would do
Improve the physical security of the shipment: more armed guards, better insurance, safer routes. It fails because the risk does not shrink with scale — every shipment of silver along a bandit-thick route is still its own independent gamble, so hiring more protection only raises the cost of moving money without ever removing the exposure, and a large enough loss can still wipe out a single shipment regardless of how many guards rode with it.
what they saw
Lei Lutai saw that a merchant did not actually need his specific coins to arrive at the destination — silver is fungible, so any equivalent amount from the receiving branch's own reserves would do just as well. What needed to travel safely was not metal but a verifiable claim on metal, and a claim is information, which can be protected by making it unforgeable rather than by guarding it physically.
the move
Rishengchang, converted from a Pingyao dye firm by owner Li Daquan and manager Lei Lutai, let a merchant deposit silver at one branch and collect the same sum at another branch hundreds of miles away on a paper draft (汇票) — no coin ever left the vault. The draft's amount, date and denomination were written not as plain numerals but in cipher: fixed poems stood in for the months, the days and the digits, and the code was changed periodically so a forger copying an old draft's wording would compute the wrong sum.
why it works
Replacing physical shipment with a paper draft eliminates the robbery risk outright, since there is nothing of intrinsic value in transit to steal — a thief who intercepts the courier gets a piece of paper, not silver. That trade creates a new risk in place of the old one, though: paper can be copied or altered in ways a chest of silver cannot. Rishengchang closed that gap by encoding the sum, date and unit in a rotating cipher known only to its own branches, so an intercepted draft cannot be recomputed or reused by anyone outside the network — converting an unsolvable physical-security problem (bandits on the road) into a solvable cryptographic one (a shared secret between two trusted parties), which a distributed branch network can maintain far more cheaply and reliably than armed escorts ever could.
the payoff
The piaohao model Rishengchang pioneered grew into a national network of roughly thirty houses and several hundred branches, reaching Russia, Mongolia, Japan and Southeast Asia; Rishengchang itself operated for over a century, surviving into the 1910s before finally closing in 1932, and the format only lost ground once the telegraph and modern foreign-style banks offered faster transfers.
where it breaks
The model depends on the receiving branch genuinely being able to pay out from its own reserves — if the network's liquidity runs short, drafts stop being safe claims and start being IOUs the house cannot honor, turning a solved theft problem into an unsolved solvency one. It also depends on the cipher staying ahead of forgers: a key that leaks, or is rotated too rarely, lets a copied draft be cashed fraudulently, which is exactly the failure mode the scheme exists to prevent. And it requires real trust between distant branches that cannot verify each transaction with each other in real time — a trust that has to come from common ownership, reputation, or an enforcement mechanism strong enough to survive the occasional dispute, since the entire system collapses if branches stop honoring drafts they cannot independently confirm.
what came after
Rishengchang's draft-and-cipher model became the template for the entire piaohao industry, which financed trade — and for a time, significant government transfers — across Qing China until the 1911 revolution and the rise of modern banks ended the trade.
references
- [1]The Shanxi BanksNational Bureau of Economic Research (Working Paper 15884), 2010nber.org
- [2]Wall Street of the empireChina Daily, 2022chinadaily.com.cn
- [3]RishengchangWikipedia, 2026en.wikipedia.org