From June 12 to December 31, 1968, the New York Stock Exchange closed every Wednesday. Not for a holiday, a panic, or a war. The market was booming. The clerks behind the trades could not keep up with the paperwork, and the only available remedy was to stop trading one day a week so they could catch up.
The arithmetic of a paper trade
Volume had climbed to roughly 13 million shares a day, with some sessions pushing near 17 million, against a back office designed around the 3 million or so a day it handled in the early 1960s. That is a four-fold increase in a decade, absorbed by a settlement process that had barely changed since the 1920s.
Every trade meant a physical stock certificate, signed and hand-delivered within a few days. Runners carried certificates between brokerage houses in lower Manhattan. Clerks matched them to trade tickets by hand, re-registered them, and mailed them out. Miss the window and you had a “fail,” a trade that had happened economically but not legally. Fails compounded, because a fail on the way in usually caused a fail on the way out.
It stopped being a clerical problem
Certificates were misfiled, mailed to the wrong address, or lost outright. In the confusion, hundreds of millions of dollars in securities were stolen, some of it by organized crime working the mailrooms and back offices directly. Firms genuinely did not know what they owned or owed.
Turnover made it worse in a way that will be familiar to anyone who has run an operations team. Clerk attrition ran near 50% a year, so the people who knew how the exceptions worked kept leaving, and the firm kept retraining someone new into a process that only existed in the heads of the people who had just quit.
Then it turned structural. When you cannot reconcile your own books, you cannot demonstrate your own capital adequacy. By the end of 1970, more than 100 NYSE member firms had been liquidated or absorbed, close to a sixth of US brokerages. Congress responded by passing the Securities Investor Protection Act of 1970 and creating SIPC, because customer assets had been put at risk by bookkeeping.
So the busiest market on earth spent half a year closing one day a week, and a sixth of its member firms disappeared, over filing.
Hiring more people was not the fix
The industry tried the obvious thing first. Firms hired aggressively into the back office, ran overtime, and added shifts. It did not work, for a reason worth naming precisely: the process required a human to move a physical object for every transaction, so capacity scaled linearly with headcount while volume scaled with market enthusiasm. And each new hire arrived into a 50% turnover environment with no documented process.
Shortening the trading week was the same category of answer. It did not raise capacity. It rationed demand to fit the capacity that existed.
What actually ended the crisis was refusing to move the paper at all. The exchange's Central Certificate Service began immobilizing certificates in one place and settling changes of ownership as book entries, a ledger update instead of a courier run. That effort became the Depository Trust Company in 1973, and electronic settlement made the whole class of problem go away.
The certificates did not get processed faster. They stopped needing to be processed.
The pattern, and where it sits today
Every operations crisis of this shape offers three responses. Ration the demand, which is what closing Wednesdays did. Add people, which fails whenever volume can grow faster than you can hire and train. Or remove the step, which is the only one that changes the ceiling.
The uncomfortable part is that the third answer is always available years before anyone reaches for it. Book-entry settlement was not a 1973 invention waiting to be discovered. It was a known idea that the industry adopted only after a market-structure emergency and roughly a hundred failed firms made the status quo untenable.
The modern version is quieter but recognizable. Account transfers rekeyed between a custodian and a CRM. Reconciliations held together in a spreadsheet. An advisory firm absorbing a book of business and discovering that integration is manual, so every acquisition triggers a hiring scramble. Nobody is closing on Wednesdays. Firms are just quietly declining growth they cannot staff, which is the same decision with better manners.
The rest of the series
Four times a profession ran out of people to absorb its own routine work, and what actually ended each one.
- IBM Exists Because the Census Could Not Finish Counting. The 1880 census took until 1887 to tabulate by hand. The machine built to clear that backlog became the computer industry.
- Who Decided Six Minutes?. One managing partner started timesheets around 1920 to run his firm better. Then the measurement became the price.
- The Paralegal Was Invented to Make Legal Services Cheaper. A 1968 ABA resolution created a role to absorb a lawyer's tedious and routine detail. The role became billable.