Silicon Insider: Intel, Buy Texas Instruments

Aug. 10, 2006 — -- Dear Intel:

Buy Texas Instruments. Now. Don't wait. Don't fool around trying to cut the best possible deal. Just pay the premium and buy TI. If you have any questions, call Andy.

In one of my very first columns for this page, nearly six years ago, I asked a very simple -- but, it turned out, also controversial -- question: Could Craig Barrett run Intel?

At the time, new-CEO Barrett was being hailed as the great successor to Andy Grove. He was reputed to be Grove's equal in intellect and experience, but at the same time had none of his predecessor's fiery temperament and take-no-prisoners management style. Barrett was described as an easygoing cowboy, who would lead giant Intel into a period of consolidating the huge growth and market dominance it had earned during the Grove era.

I didn't question any of those appraisals. What I did ask was if those really were the right characteristics by which to judge Barrett's ability to run what was, at the time, the most valuable manufacturing company on earth and the cornerstone of America's high-tech leadership.

Neither Intel nor its many admirers would have any of it. Barrett, after all, had been a key player in getting the company to its then pre-eminence in the semiconductor industry. Even more important, thanks to its (tense, but industry-defining) partnership with Microsoft, Intel enjoyed a near-monopoly in microprocessors -- that most important of all the tech businesses -- that not even an idiot could lose. And, frankly, after Justice Department antitrust investigations and accusations of hardball business tactics, both customers and employees were ready for a little Pax Intel after 20 years of Andy Grove table-pounding.

Logging Intel's Mistakes

But I knew something else about Intel. A few years before that, I had written a book about the microprocessor that, at the time, was still being used in some university engineering survey courses. The book ("The Microprocessor: A Biography") was essentially two books in one: an explanation of the technology and how it worked and a history of microprocessors and the semiconductor industry.

Having covered Silicon Valley already for nearly two decades by then, I assumed the latter story line would hold few surprises for me. And yet, as I wrote the narrative, I was suddenly struck by something I'd never noticed before.

It was that Intel had screwed up an amazing number of times in its first 30 years. Indeed, it had probably made more mistakes than any other chip company during those three decades.

That certainly didn't correspond to the common perception of Intel as an army of single-minded, humorless engineers relentlessly chewing up market share in microprocessors, taking on all comers, big (Motorola) and small (Advanced Micro Devices), innovators (Zilog) and manufacturers (TI), the kindly (HP) and the ruthless (National Semiconductor) -- and beaten them all. Intel had taken the title of the Most Important Company in the World in the early '80s and had tenaciously held it against all comers ever since. Even more than HP or IBM or Microsoft, Intel had been the gold standard in tech, the company that could be depended upon to never stumble.

But in studying the history of Intel, I suddenly saw all in one place all the mistakes that I had only noticed piecemeal and had written off at the time as momentary aberrations: the company's near decision not to pursue semiconductors; the overinvestment that almost led to bankruptcy in the early '80s, the hanging on to the memory business until it was almost too late; the endless wasteful lawsuit with AMD; the falling almost fatally behind Motorola in the '70s, the Pentium Bug; the vicious straight-arming of customers in the late '90s that led to a Justice Department investigation … etc., etc.

Any one of these mistakes could have (and indeed, had) killed or crippled other great companies. Yet Intel had not only survived them all but thrived, eventually coming as close as legally possible to enjoying ownership of one of the most lucrative businesses ever devised.

What was the difference? Certainly Intel had been able to recruit top talent for three decades. And its x86 processor architecture has proved incredibly durable. Finally, the unmatched reputations of the company's two founders, Gordon Moore and the late Bob Noyce, certainly helped get the company through hard times when customers might have bailed. But not one of those factors was sufficient. No, I realized it was something else: the leadership of Andy Grove.

Was It the Man or the Company?

Grove has long been celebrated as one of the greatest business leaders of the last half century. He was even Time's Man of the Year. And yet, there has always been the sneaking suspicion that his reputation had a lot to do with the company he led, and that any moderately competent executive could successfully run a company holding the cards Intel did.

But I came to realize that the truth was exactly the opposite: Intel appeared a great company precisely because Andy Grove ran it. Every time the company stumbled, Grove was there to pick it up by the collar and kick it in the butt. That was impressive enough, but the real jaw-dropping moments in Intel's history came when it was Grove himself who screwed up … then did a complete turnaround and rectified his own mistake, even at the cost of public humiliation.

It was Grove, after all, who hung on too long to Intel's memory business -- then dumped it and devoted all his energies to processors. It was Grove who, on learning of the Pentium bug, chose to insult company customers … then turned around in 72 hours and drove the company to make sweeping fixes. And it was Grove who led the company to the brink of a Justice Department investigation … and who (unlike the foolish Bill Gates) did a quick mea culpa and probably saved his company.

There were a lot of things not to like about Andy Grove's management style. I couldn't have worked for the man for more than two days. He was relentlessly competitive, arrogant, vindictive and uncompromising. But that is what it took to make Intel the great company it was. And nobody was more loyal to Intel than Andy Grove. I once wrote something cruelly funny about Intel, and Andy didn't speak to me for 10 years. That was Andy, and that's why he was the best.

TI's History Not So Sweet

Texas Instruments was a very different story. If Intel was a good company made great by leadership, TI was a great company driven into the ground by bad management. Only tech old-timers still remember when Texas Instruments ruled the rest of the electronics universe not owned by IBM and HP. It was at TI, after all, where Jack Kilby came up with the first design of an integrated circuit -- a feat matched only by the parallel work of, ironically, future Intel co-founder Bob Noyce at Fairchild Semiconductor.

The fact that TI would have the idea, but it would take little Fairchild to execute it, was an early glimpse into the weak soul of TI. A dozen years later, it would happen again: This time, though, TI was hard at work on a microprocessor design. It would be Intel that would be first to market -- leaving TI a perpetual also-ran until it finally dropped out of the race.

TI also pioneered the four-function calculator -- only to see the Japanese over run the market. Meanwhile, it was HP that invented the high-end scientific and business calculator markets. TI came late to that market, and then, using the absurd learning curve theory of pricing, so price-bombed the market in pursuit of share that it drove the profits right out of the business.

Looking back, Texas Instruments always seemed a vast company with infinite talent and resources but without an ounce of greatness. By the '80s it was exhibiting all the classic symptoms of bad management: labor problems, poorly conceived products, also-ran positions in key businesses, sinking market share, a disappearing company profile. By the early '90s, the company was reduced to suing competitors and partners over the early chip and processor patents it owned. It was said at the time that the largest profit center in all of TI was its legal department.

If you had looked at the two companies in the mid-1990s you would have concluded that Intel was invincible and Texas Instruments was doomed.

But nothing ever stays the same for long in high tech.

A Rebirth Fueled by Cell Phones

At TI, there was one area of technology in which the company had always enjoyed leadership: digital signal processing. It wasn't much of a business: For decades, DSP chips were primarily used to handle analog signals coming into digital computers from things like instruments and sensors.

Then along came cell phones. And wireless. And the whole communications revolution. While nobody was noticing, wounded and dying Texas Instruments found itself sitting on one of the hottest technologies of the age -- and given a new lease on life. These days, with annual revenues of $15 billion and market leadership in DSP (and other related technologies like analog-to-digital conversion, power management, etc.), TI is one of the hottest niche players in all of tech.

During those same years, Intel looked unstoppable. Its partnership with Microsoft continued to rule the PC world. The "Intel Inside" campaign was one of the most successful in advertising history. And the generations of industry-defining Pentium chips appeared like clockwork.

But there was one crucial difference: Andy Grove had retired. Under Barrett, Intel was still a well-run machine, but -- to the relief of everybody -- it no longer behaved as if its very existence was at stake every single day, as if it was surrounded by enemies, that, to use Grove's famous book title, "Only the paranoid survive." Now when the company made mistakes -- which it continued to do, letting perpetual follower AMD not only catch up, but take the lead, blowing $10 billion on a failed thrust into communications, not recognizing the devastating threat in the Far East from Samsung until it was almost too late -- there was concern and the announcement of new corporate initiatives. But it was nothing like the Grove era, when there would have been heads on poles, rallies calling for employee sacrifice unto death and key customers being dangled out of windows by their ankles. Instead, Intel exhibited something I'd never seen before in the company: fear.

Before he retired, Barrett finally found his inner tough guy. But he never learned to be vicious. And he never learned to be relentless. His successor, Paul Otellini, is a very smart man and a superb executive. But he too has yet to show that he knows how to play for keeps, that he is willing to whip Intel back up on its bloody feet the next time it falls. Because that is what it will take to beat Samsung.

Buying TI Will Take Guts

The first step on Intel's long march back to dominance has to be a move so gutsy that the whole world catches its breath. Perpetual antagonist AMD has already made such a move -- buying a company, ATI, it can barely afford in order to extend its position in high-speed graphics, games, etc. That took guts. If the next recession comes early to the chip industry, AMD is going to be in a bad way. That's an Andy Grove play -- ironically made by the company he hates more than any other.

Now it's Intel's turn to bet the store. Frankly, it has little choice: If Intel continues playing with its current hand, it will be slowly crushed from below (Samsung) and above (AMD). Instead, it needs to change the rules of the game, something Grove did better than anybody. Buy Texas Instruments and it gets industry leadership in DSP. Bolt that together with multicore processors and Intel can steam away from AMD and hit Samsung head on. Intel could own digital communications the way it has own microprocessors.

And, needless to say, in a beautiful alignment of history, the merger would also at last bring together Noyce and Kilby, closing the circle on the first half century of the semiconductor industry.

There is the tiniest of rumors on the street in Silicon Valley that Intel may already be pondering such a move, that it may even have made preliminary contact with TI -- but that Intel is hesitating because the price may be too high.

Well, of course it's too high. This would be one of the biggest mergers in American business history. It reshuffles the deck in tech. It's supposed to make your knees turn to jelly. But this is what great, $40 billion companies do. This is what Andy Grove would do.

Intel, buy Texas Instruments.

This work is the opinion of the columnist and in no way reflects the opinion of ABC News.

Michael S. Malone, once called the Boswell of Silicon Valley, is one of the nation's best-known technology writers. He has covered Silicon Valley and high-tech for more than 25 years, beginning with the San Jose Mercury News, as the nation's first daily high-tech reporter. His articles and editorials have appeared in such publications as The Wall Street Journal, the Economist and Fortune, and for two years he was a columnist for The New York Times. He was editor of Forbes ASAP, the world's largest-circulation business-tech magazine, at the height of the dot-com boom. Malone is best-known as the author or co-author of a dozen books, notably the best-selling "Virtual Corporation." Malone has also hosted three public-television interview series, and most recently co-produced the celebrated PBS miniseries on social entrepreneurs, "The New Heroes." He has been the ABCNEWS.com "Silicon Insider" columnist since 2000.