1967 and 1968 were probably the most stressful and busy years of my career. I was in charge of my firm's operations and financial departments (known as the 'back office'). NYSE trading volume was very high-running at approximately 10m shares per day, up from just 2m per day in 1950.1 The operations departments of many broker-dealers were in deep trouble caused by the high trading volumes. 'DKs', (later renamed 'QTFs') were endemic, and a number of NYSE firms lost bookkeeping control and were forced to close or merge. My workday was 12-14 hours long, including weekends.
To deal with this crisis, from August 8 through August 18, 1967, the NYSE closed at 14:00 instead of the usual 15:30. Then, from January 22 through March 1, 1968 the Exchange closed at 14:00, again because of the back office work load. That step did not fix the problems. Things got so bad that from June 12 through December 31, 1968 the NYSE went to a four day week, being closed on Wednesdays or regular holidays. Wall Street's problems became nationally known as the 'Paperwork Crisis'.
It was in this environment, with serious operational problems originating in large part from errors and delays on the trading floor, that a joint team from the NYSE and IBM Corporation, after working on a study for several years, finally produced a 100-page report urging development of an 'advanced integrated trading system' for the NYSE and delivered it to the NYSE's top management.
This report recommended a complete revamping of the trading system. Instead of face-to-face trading, the plan was to use computers and trading terminals with the following objectives: to speed up executions, handle increased volumes, reduce errors, improve surveillance and increase cost-effectiveness.
There was no publicity about this study and report. Instead, upon delivery the report was immediately buried by management, and not even shown to the NYSE's Board. When The Wall Street Journal reported on the study a year later, it was immediately dismissed by the Exchange as unrealistic and as merely technicians' dreams. In a memorandum to members the day after the article appeared in the Journal, that newspaper was harshly criticised for reporting on a story which NYSE had already explained to them that the study had been rejected and wasn't worth writing about.
Here are a few excerpts from the rejected report:4
Study background
The initial Joint Study Agreement between IBM and the NYSE originated in 1961. The present study, formulated in September 1967, is merely an extension of the original agreement. The general purpose of these agreements has been to investigate the possibility of technological improvements in the operation of the NYSE.
It went on to state:
...we conclude that it is essential for the Exchange to begin now a broad development and implementation program leading to an advanced integrated trading system. We believe that this is the most practical way to deal with the great increases in trading volume that the Exchange will continue to experience. Continued expansion of the present cumbersome and largely manual floor system has become increasingly difficult as its saturation point is approaching.
The report went on to address the benefits of the trading system it recommended:
Implementation of the proposed integrated trading system will enable the NYSE to continue to confine all trading activities to one physical site. Some advantages of this are:
- It facilitates means of providing adequate, reliable fallback, back-up and maintenance procedures in the event of failure of component systems;
- It should prove to be more economical and efficient, especially in the areas of personnel and communications costs;
- It maintains the image and identity of the Exchange in the minds of the general public;
- It does not create any conflicts with constitutional rules of the Exchange especially those relating to on-floor and off-floor trading.
Within the one site, members will trade from in individual or member firm customised offices or stations rather than trading in large, open areas like existing floor areas. Trading will be performed 'remotely' in the sense that members will use various data communications devices to communicate with one another. Furthermore, members will trade functionally as they currently trade and under essentially the same rules and regulations as that exist today
The report goes on to state:
After the Specialist is trading full-time on the integrated system, he will be relocated to an office away from the post where he will continue trading. Another Specialist will then be started on the integrated system as described earlier. As Specialists move from posts to offices, other Specialists on the Floor will be provided with the facilities of integrated trading systems...At some point in this procedure, floor brokers will also move to individual offices. When all specialists are on the integrated system, all floor brokers will be located at offices.
The report concluded:
...it is the unanimous opinion of the Joint Study Team that, although manual methods are possible, locked-in trading by a method other than by some type of trading console cannot be achieved without cumbersome procedures, an exorbitant number of personnel, or disruption of the trading process.
And finally:
The study concludes that, consistent with current technological trends, such an integrated system represents the best long-range solution to floor trading requirements. It recommends that the Exchange officially commit itself to the development of such a system.
As the late Yogi Berra would have succinctly stated about the changes now being made in 2006, 'This is like déjà vu all over again!'
In 1968, the Securities and Exchange Commission's efforts were focused on improving the back office operations. However, they failed to address methods of improving the mechanics of trading with the increased volumes. Today, thanks in large part to the efforts of the industry and the SEC, operational problems are now relatively minor; total trading volumes running above 3bn shares per day are handled with alacrity and ease. The advent of the digital computer and sophisticated telecommunications changed the operational landscape dramatically.
Not so for the trading systems. They remained almost unchanged. However, as late as 2004, the driving force for improved automation of trading on the NYSE finally arrived. That force was the existence of competing electronic trading systems with the NYSE's floor, and SEC proposals (made 30 years late).
There is, of course, no guarantee that the NYSE-IBM study recommendations could or would have worked as the study's authors believed. What was sad - but not surprising - was that a high cost, professional team spent years developing the plan, only to discover that the Exchange's management apparently were so frightened of the idea that computers and telecommunications might supplant or replace the century and a half systems of face-to-face trading, and also would take from the specialists and floor brokers their competitive advantage over investors they enjoyed for so long.
This mind set, however, continued for more than three decades after 1968. The NYSE successfully expended millions of dollars on lobbyists to thwart the plain language of the Securities Reform Act of 1975, which called for the SEC to facilitate the appropriate use of modern technologies to improve securities trading. Instead, they offered pseudo-automation, such as the Intermarket Trading System that William Schreyer, the president of Merrill Lynch, once described under oath before a congressional committee as follows:
...The Intermarket Trading System, or ITS, which links the New York with some regional exchanges, is a communications device, and nothing more. It is as far from the concept of an automated, efficient marketplace as a tom-tom from a communications satellite.5
Frequently, at public hearings and at congressional testimony, the Exchange's spokespeople and allies would warn that changing from an auditory auction to a visual one would never work, and that any attempts to use computers to execute trades would destroy the U.S. stock market.
In the meantime, others started work on automating the trading process. The Toronto Stock Exchange implemented 'CATS', a computerised system. Instinet was formed. Merrill Lynch proposed an automated market in 1975. The computerised Cincinnati Stock Exchange started trading.
Around the world, old and new markets sprung up, none using the NYSE's model. Trading floors started to vanish, both in securities trading and futures trading.
It was only after the Commission approved Regulation NMS that required efficient trading using automated markets that the Exchange's new management was forced to make the watershed decision that attempts to bring it into the 21st century.
However, the NYSE, in order to convince its members to convert the Exchange to a public company, offered what they term a 'Hybrid Market'.
Their plan was to operate two different trading systems simultaneously. On the one hand, the NYSE would be the owner of the fully-electronic Archipelago ECN and also a semi-automated traditional floor trading system.
As the London Stock Exchange discovered in 1986, operating manual and electronic trading systems in the same securities at the same time did not work. After the 'Big Bang', as the experiment was known, the floor disappeared for trading almost immediately.
Is the NYSE's 2006 attempt to emulate what failed in London going to work? Or will it be the Arca portion of the merger that will win the day? Stay tuned!
Notes
1 By contrast, on February 24, 2006, NYSE volume was 1,436,451,940 shares.
2 'DK' stood for 'Don't Know', a trade made on the Exchange floor that had an incorrect or missing counterparty.
3 A 'QTF' was exactly the same as a 'DK', and stood for 'Questioned Trade Form', and was renamed because the management of the NYSE did not want to admit that trades that did not compare properly were commonplace.
4 All quotations are taken from the report of the NYSE-IBM Joint Study, May 1968.
5 Joint Hearings before the Subcommittee on Oversight and Investigations and the Subcommittee on Consumer Protection and Finance of the Committee on Interstate and Foreign Commerce, House of Representatives, September 21, 24 and 25, 1979, Serial 96-89, US Government Printing Office, Washington, 1979.
