IBM is making no bones bout it. The company’s in deep trouble, and will be hurting for rest of this year. In describing its principal difficulty, IBM confirmed the somewhat speculative news reports – first aired here as long ago as February 18 – and Wall Street rumours that had made the rounds for months. The gist of these stories was that IBM had announced a modest renovation of its 3090E mainframes – the 3090S line – but was unable actually to deliver the machines to customers. On Friday, March 17, IBM made it official. The company said, in obfuscatory fashion, that it had had a problem making as many 3090S systems as customers wanted. Consequently, the manufacturer added, its earnings for the quarter would be lower than expected. Incidentally, explained the manufacturer’s spin doctors, profits for the rest of the year would be below expectations, too. IBM shares fell nearly nine dollars between Friday morning and Monday’s market close, wiping out more than $6 billion of paper wealth around the world. Combined with reports in the US, UK and across Europe that inflation was back in style – prices of most IBM computers, for instance, were being increased by 5% in the US and 3% in the UK – the impact on equity markets was instant and quite severe. And the full effects of IBM’s statements may yet be forthcoming. IBM’s earnings forecast is particularly bad news when one considers just how Wall Street analysts get their estimates in the first place. The most respected and influential stock watchers don’t merely report the numbers that come up on their spreadsheets. An IBM analyst will let IBM know what his quarterly profit numbers are and then expect IBM to drop some subtle hints if the company feels a revision is in order. IBM doesn’t say, Raise it 32 cents, idiot!, or You’re a dime too high. But the company would tell an analyst that, You may be somewhat optimistic. This, as part of a continuing dialogue, has a specific meaning to a particular analyst; it might, for instance, mean that the analyst is 5% to 10% above IBM’s own secret forecast. So when IBM said that analysts’ profit expectations were excessive it was, in effect, admitting that its own projections had gone awry. Rather than express their gratitude to IBM for its taking blame for their mistakes, the analysts instead turned on Big Blue. This is natural enough. For several months to come, the stock-watchers will actually have to do their own work. More importantly, many analysts now face a new risk: the firms that pay them hundreds of thousands of dollars may have discovered that some of the more visible voices on Wall Street emanate from heads containing no more grey matter than those of healthy squirrels. Despite its appearance of being under fire, IBM has in fact gotten away with quite a bit so far. The company’s disk business is in dire straits, but not a word about this has appeared on the news wires. Quite a few people suggest that a significant number of the AS/400s shipped by IBM are languishing in the warehouses of remarketers. The OS/2 operating system, IBM’s heroic effort to wrest control of the PC software market back from Microsoft and a host of other barbarians, has been selling as well as snowshoes in the Sahara. Various less important products, like the RT and the 3480 mainframe tape subsystem, also have brought disappointment to IBM. The bright side of these mishaps – a reduction in overhead as hundreds of careers meet the corporate guillotine – hasn’t yet been appreciated by investors. Nor has the likelihood that IBM will have some good news tomorrow, when it is the guest of a huge convocation of stock pickers in Boston. Users, however, already seem to sense the company’s vulnerability. Despite IBM’s protestations that its problems stem from low chip yields, not a lack of orders, many mainframe customers seem to be getting hefty discounts. This is particularly apparent at the largest, savviest shops – the ones that will plug in an Amdahl mainframe or a used IBM machine if Big Blue trys to be too clever and overplays its hand. The toll mou
nts Meanwhile, neither IBM nor the users of its gear – nor the purportedly great minds of Wall Street – have examined the roots of IBM’s woes. IBM may have accelerated its 3090 announcement plans when Amdahl first said it could build 100 MIPS mainframes. Instead of simply announcing one additional Amdahl-baiting machine – the 3090-600S – which it could have delivered no matter how low its chip yields were, IBM decided to unveil the whole 3090S product line. Presumably, the company’s engineers agreed they could meet the target, an assertion obviously made under great pressure and in an atmosphere of fear, uncertainty and doubt concerning their careers. So far, IBM has survived its inability to keep its promises. But the toll mounts. It includes the 3990-3 disk controller that is about a year behind schedule, the performance problems associated with the smallest AS/400s, the death of IBM’s portable PCs and its RT headaches, the initially cool reception accorded OS/2, the inappropriately high prices of 3380 disks, the sad case of the 9370, the company’s chaotic attempt to cope with the Unix craze, and its unexpected endorsement of Japanese mainframe standards that might undermine its own. Eventually, somebody might question whether the company’s leaders are in control of their situation.
