Chapter 7 - Data Communications: Market Order 1973-1979
7.11 Codex and Motorola 1977-1978
In mid-1976, amidst the continuing challenges of ramping production of the L-series modems, finishing the statistical multiplexer and wondering if business conditions would ever improve, Carr had a visit from Keith Bane, director of corporate development of Motorola, proposing Motorola acquire Codex. Carr recalls:
Bane gave me this pitch why we ought to be bought by Motorola, and I basically threw him out. Their plan was that they were going to buy five or six companies for $100 million and be in the data com business, and they had to do that because they had no in-house expertise. I told him he couldn’t find five or six good companies, and if he did, he couldn’t manage them, and he couldn’t do it for $100 million anyway. Therefore he didn’t know what he was talking about, and we had, in our long range plan: ‘we will not be acquired by anybody.’ And I showed him the door. On the way out he said: ‘If I come up with a good plan, can I come back,’ and I said: ‘Yeah.’
The chilly reception Carr gave Bane masked his own gnawing concerns of how best to navigate Codex through an increasingly complex environment. Gone it would seem were the days of almost gentlemanly competition among a handful of well-known-to-each-other firms. Now each week brought new surprises, and fateful signs that the lumbering AT&T and the voracious computer companies saw data communications as another means to achieve market domination. Was there a future for independent data communication companies, or were they destined to be swallowed up by the larger, more diversified firms? Carr decided to engage his senior managers in serious reconsideration of assumptions long held inviolable.
A dispassionate evaluation of modems came first. Fortunately, they had seen the implications of semiconductor technologies for modems and acted. To believe the press, every high-speed modem competitor had LSI modems under development, if not already announced.70 The first to announce an LSI modem, Paradyne, had set the tone of coming competitive dynamics with aggressive pricing. (Although the August bankruptcy announcement of Datran might be the undoing of Paradyne and permit a return to more traditional market behavior, or so Codex management hoped.) Semiconductor microprocessors introduced another uncertainty. For example, Milgo had recently introduced Fairchild microprocessor-based modems.71 Would the reprogramability of microprocessors prove more important than the speed and costs of LSI custom designed circuits? Another new development, initiated by the Codex-spinout Intertel, emphasized the integration of network control and diagnostics into modems. Then there was the advent of multipoint modems, modems that allowed multiple modems to share the same telephone circuit, thus dramatically cutting communication costs. How would Codex fare in the face of such divergent modem developments?
As confused as the future of modems appeared, it seemed comparatively certain against the nightmare of deciding what products to offer after multiplexers. Multiplexers were an obvious choice as they made modems much more cost effective. Many thought computer terminals logical products to sell as almost every data communication application involved connecting remote terminals to host computers. However terminals, historically decidedly low-tech, were becoming intelligent which probably meant rapid product development cyles compared to modems and multiplexers, competition was fierce – every major computer company, including AT&T, sold terminals as did countless independent firms – and servicing mechanical products in the field was a real headache. If not terminals, then what products could be sold to the computer-end of circuits? Yet just as Codex had deferred innovating a front-end processor because of the hurdle of developing and maintaining host computer software, any host-end product ideas always seemed to involve a massive software effort, a competence Codex did not have.
More threateningly, the computer companies now sold data communications products and were introducing sophisticated networking software products. IBM had started the trend in 1974 with SNA. In 1976, they significantly enhanced SNA, to include support for multiple computers. In reaction, every major computer company announced similar products, including the minicomputer manufacturer DEC with DECNET in 1976. If the challenge of developing host computer software had seemed daunting before, it had now become a seemingly impossible undertaking. The question facing the data communication companies had suddenly switched from one of: How to achieve growth by interfacing products to computers? to one of: How not to be locked-out by the computer companies?
Any contemplation of the future had to include the nemesis of all data communication companies: AT&T. The DAA (Data Access Arrangement) was going to become a relic of the past, although not until June 1977. Not to be counted out, however, Pacific Bell had publicized the idea of “connecting blocks,” another way AT&T could charge for interfacing foreign attachments to their network – the idea would die a deservedly quick death. But more shenanigans could not be dismissed. In addition, AT&T had the technological and financial muscle to wipe out the independent manufacturers if it ever set its mind to doing so.
Then, in a stark reminder of the unforgiving nature of markets and the volatile state of data communications, on November 8, 1976, Applied Digital Data Systems (ADDS) announced a take-over attempt of Milgo. Could Milgo escape the embrace of the unwanted suitor? Carr and Codex management had to wonder if they were to endure the same fate only with Motorola.
In December, Bane returned as promised and presented a new plan of acquisition. Carr considered the new plan only marginally better than Motorola’s prior plan and bluntly said so. Bane would not be denied, however, and persuaded a curious Carr to meet Robert Galvin, Motorola’s CEO. Carr, having reviewed Codex’s options with management, and stunned by the victimization of Milgo, realized he felt far more disposed to the idea of being acquired than he had assumed. He had to chuckle thinking of the trials of the past year: they were enough to cause anyone to question why they would want to be President of a young technology company. Carr recalls:
Then, here came Motorola and said: “If you’re buried in Motorola, you’re not going to be visible. You don’t have to perform every quarter. We can fund you. Our interest is in making you the foundation of a new business which we’re willing to fund. Your interest is in dominating your marketplace. With us, you can do more than one thing at once. We started thinking about how much capital we could raise, and we thought we could raise a lot of capital, but we didn’t see how we could manage the P&L. We used to have an analyst meeting every quarter, and we had analysts who stood up and were unsatisfied with 36% pre-tax. They’d say: ‘You’ve been doing 36% for six quarters now. When are you going to improve it?’ You wanted to climb up on the guys chest and take a hold of his tie. We were literally in the mode where we were almost doing nothing wrong and they were on us about things we were doing right and why weren’t they better. And the board, believe it or not, was giving us the same routine. ‘Why don’t you crank up earnings per share?’
So the insiders, Storey, Forney and I, finally concluded that A, we had had enough of the board; B, that we could probably fund both the front-end and the terminal launch inside of Motorola; C, they were our kind of people. We spent a lot of time on that, in terms of ethics and the way they went about things, culture. Lastly, we would not only be autonomous, but we would be the core. I had told Galvin all the horror stories of Computer Control being acquired by Honeywell at the time it was just passing DEC, and dissolving it over the course of the next three years. So he told me that if we did this thing, that I would be in charge of managing the acquisition, not somebody at Motorola, because they didn’t know anything about it. So we finally decided that, from a strategy point of view, we believed we could advance the fortunes of Codex more significantly being buried in Motorola and drawing on their resources than we could by being a free-standing company and having to dance the quarter-by-quarter tune.
On February 7,1977, Motorola announced an agreement in principle to acquire Codex in an exchange of stock valued at more than $80 million. On February 23rd, Racal acquired Milgo. The two leading independent data communication firms, and long-time foes, now had to learn if being small entities in much larger corporations gave them the resources and freedom to extend market leadership. Or would they become data communications-also-rans in a market that had suddenly become much more challenging?