The Oil Giant That Went Into Typewriters: Even at Xerox's Size, It Was a Drop in the Barrel
In 1978 Exxon merged its word processors, smart typewriters and fax machines into Exxon Information Systems to take on IBM. What it lost wasn't the technology but its own scale: matching Xerox would have been one drop in the barrel.
The argument · tap a timestamp to hear it
Oil stopped being a growth business, and that forced diversification
In 1958 Jersey Standard sold more oil than ever in its history, yet profits were only a third of what they had been a decade earlier, and Forbes said a revolution had taken place in the world oil economy. In 1960 CEO Monroe Jack Rathbone ordered a reassessment of the company's investment philosophy, and the board put finance executive David J. Jones in charge of the report. The Jones Report concluded that oil and gas were no longer a growth business and that future returns on oil and gas opportunities would only get worse — but it did not argue for exit. It recommended more caution in oil and gas investment while opening new revenue lines. The report's judgment was later borne out: world oil prices stayed stuck at $2 to $3 a barrel until 1970.
The chemicals expansion ran too fast, and returned only 4%
The most obvious direction for diversification was chemicals. Esso Chemical was then just a byproduct division, with 1959 revenue of $250 million, making basic chemicals like isopropanol, acetone, and ethylene and propylene. But if petroleum feedstock was going to stay cheap for the long term, why not make higher-value-added products? The board set up a standalone subsidiary and raised the annual investment budget from an average of $53 million in 1961 to 1963 to $257 million in 1964 in one jump, ordering fertilizer plants from Central America all the way to Malaysia, Pakistan, Greece and Canada. Rathbone himself told Forbes that everyone was rushing into chemicals, that this would lose a lot of money for some companies, and that it would not be them. The result: Esso Chemical's return on net assets was only 4%, thermoplastic and fertilizer capacity went into oversupply, and at the end of 1966 management was forced to pull back.
Ten candidate sectors, and only information technology was left
Jersey Enterprises, set up in 1963, was responsible for exploring new investments. It first picked out ten broad categories that could grow into $100 million companies, looking both inside and outside. Most became jokes. In 1964 it paid $37 million for American Cryogenics, hoping to get industrial gas technology, and gave up four years later because the technology was not competitive. In 1965 it partnered with a Nestlé affiliate in Guatemala to grow bacteria on hydrocarbons and produce a protein food, WO-100; the finished product was a bland white powder that reportedly performed poorly when fed to chicks unless extra vitamins were added. Binding soil and aggregate into building blocks with petroleum residue cost about the same as cement blocks but could not be shown to be more durable. Medical instruments, nuclear power and fuel cells were also eliminated one by one, and the ten directions were slowly narrowed to a handful — one of which was information technology.
Vydec worked, but it was a rounding error next to oil
Vydec was the first success among Exxon's IT investments. It began as a terminal project inside Hewlett-Packard; after the project was killed, its head, Patrick P. DeCavaignac, was allowed to take it out and start a company, building a word processor with a screen and floppy disks in a basement — one of the earliest products with a floppy drive. After several VCs turned them down, they reached Jersey Enterprises through a friend of a friend. Exxon studied it for a long time and in early 1973 put in $500,000 for 40%, eventually holding more than 80%. With Exxon behind it, Vydec could build its own manufacturing capacity and a large sales force, reaching about $24 million in 1977 revenue and opening more than 50 sales offices. But that same year Exxon did $48.6 billion in oil alone — $24 million next to that is peanuts.
Too much money and too many directions threw Zilog into disarray
In 1975 Federico Faggin and Ralph Ungermann left Intel to found Zilog, with Exxon as the sole investor. The early-1970s stock crash had dried up the venture market; Faggin recalled that all technology investment in 1975 totaled just $10 million, and Exxon was the only player in town — Faggin also acknowledged that without Exxon, Zilog probably would not have gotten off the ground, and the money did help it build a fab. But Exxon's money was a double-edged sword. Former employee Bernard Pueto said the oil giant gave them too much money and too many directions, making Zilog do things it should not have done, while its young managers lacked the experience to judge when to push back. Faggin later spent more time in New York fending off Exxon than in California with customers, and his relationship with Ungermann broke down over it; Ungermann left.
Analysts believed the office business could surpass oil
At the end of 1978 Exxon merged Vydec, QWIP and Qyx, along with a dozen other investments, into Exxon Information Systems. Sykes said the operation did $100 million in 1978 and $200 million in 1979, with the goal of becoming a major player in office equipment within three to five years. A Yankee Group analyst said in 1980 that the business could eventually match or even surpass their oil business, and predicted that the entire IT market would be worth $150 to $200 billion by the end of the 1980s, with Exxon taking 10%. The arithmetic on the other side was blunt: Exxon's 1979 revenue was $84 billion, so even if the office business grew as big as Xerox, it would add only $7 billion — one drop of oil falling into the barrel. And EIS was born with 6,000 employees and a 40-page organization chart.
A typewriter caught fire in Exxon's own New York headquarters
Faggin recalled the company's decision speed: every decision had to work its way up from the very bottom, and in the end senior vice president Ben Sykes said yes or no. The product level was just as embarrassing. The second-generation Qwip fax machine took six years to develop and died on launch — the paper feeder jammed. Vydec was shut out of U.S. government orders over security concerns. A Qyx smart typewriter caught fire inside Exxon's New York headquarters. The Houston marketing department preferred Wang, and after being forced to switch to Exxon equipment, switched back a year later. Exxon had bought a whole generation of old-style word processors just as the PC revolution hit, and it had neither the structure nor the culture to cope with a fast-changing, talent-centered information technology industry.
After twenty years of trying, they were still oilmen
Exxon's identity as an IBM competitor ensured that IBM would not choose Zilog's Z80 to drive the IBM PC — Faggin later confirmed this was indeed a factor. In 1981 Exxon cut staff twice, taking EIS from 6,000 people down to 4,000; that same year the IBM PC arrived and further obsoleted EIS's already outdated product line. In 1984 the company finally admitted it did not work and began selling off business lines and shutting down the rest; according to a former manager, the remaining QWIP fax machines were hauled to a junkyard and crushed with a bulldozer. Zilog was the last of the portfolio to be sold: a late-1980s management buyout backed by Warburg Pincus at about 4 times free cash flow; it went public in 1991 and was acquired in 1997 for $527 million, a 12x return on the $17 million investment. In 1993 Lee Raymond became CEO and reaffirmed that they were oilmen, and nothing else.
In their own words · checked verbatim
Everybody and his brother, for example, seems to be making polyethylene. But this dashing into chemicals is going to lose a lot of money for some companies, because petrochemicals are not a guaranteed gold mine ... And it won't be us.
Rathbone9:18
This is a running theme with all of Exxon Enterprises' little projects. How does any of it compare to oil?
gave them too much money and too many directions. It got Zilog into things they shouldn't have done, and their young managers didn't have the experience to know when to push back.
Bernard Pueto17:38
It seems likely that the [office system] business they are building now will rival, if not surpass, their oil business ultimately
Yankee Group analyst19:38
For Exxon, that's a drop of oil into the barrel.
If God himself came down in a junior businessman's uniform, worked very hard for three years, and got very lucky, he'd still have a tough time pulling those companies together
John Cunningham20:39
They were oil men. Nothing more.
Figures
| Jersey Standard 1958 revenue | $7.5 billion, second on the Fortune 500 | 4:11 |
| World oil price (until 1970) | $2-3 a barrel | 6:12 |
| Esso Chemical return on net assets | 4% | 9:18 |
| Price paid for American Cryogenics | $37 million | 11:23 |
| Vydec 1977 revenue | about $24 million | 13:27 |
| Exxon 1977 oil revenue | $48.6 billion | 13:27 |
| Exxon 1979 revenue | $84 billion | 19:38 |
Glossary
- Jersey Standard
- The entity left after the 1911 breakup of Standard Oil; renamed Exxon in 1972
- dedicated word processor
- A typewriter with a video screen and data storage that let you edit text before printing
- Directly Reduced Iron (DRI)
- Using a hot reducing gas to turn fine iron ore into metallic iron, skipping the sintering and pig iron steps
How to listen
Founders and investors interested in giant-company transformation and the failure of internal innovation at large corporations — especially anyone about to push a new business line inside a mature company.
The 1911 breakup of Standard Oil and the early corporate history — skip straight to 5:12.