Media Unspun catalogs reactions to the latest Microsoft decision:
Media Unspun: ...In his 1998 book "Barbarians Led by Bill Gates," the lead [Microsoft] engineer on the Pen Windows project, Marlin Eller, wrote of bucking up a discouraged co-worker after Pen Windows shipped but was not selling many copies. "Pen Windows was a winner. We shut down [the] GO [Corporation]. They spent $75 million pumping up this market, we spent $4 million shutting them down. They're toast. That company is dead. We did our job."
Let me parse this for you. The (unnamed) coworker had thought that the point of working on the handheld pen-computing operating system was to create a new industry: to add what we now call "Palm-like" hand-held computers to the set of useful electronic devices. The unnamed coworker was depressed because after manyyears of hard work, Pen Windows was a dog that nobody seemed to want to buy and Pen Windows's major competitor, GO, had gone bankrupt in large part as a result of its customers' decisions not to buy the (well thought-of) GO but to wait until Microsoft's Pen Windows came out before deciding on what hand-helds to buy, if any.
Martin Eller points out to the unnamed coworker that--from the perspective of Microsoft headquarters--Pen Windows was not a failure, but a smashing success. The destruction of GO computer was, in Eller's estimation, a major win for Microsoft. As long as there are no effective handheld computers, there is less competition for Windows on desktops and laptops.
Even though Pen Windows had failed in its secondary mission as a system that people would want to use on handheld computers, it had succeeded in its primary mission. By performing a "cashectomy" on GO's initial-year sales, the fact that Pen Windows was coming out real soon now had done its job in maintaining Microsoft's Windows monopoly.
From the consumer's point of view, this was a big loss. The (poorly performing) Pen Windows, by destroying GO, left consumers without handheld pen computers offering acceptable performance for about five years, until the emergence of Palm.
Is it really the case that Microsoft's principal goal in spending $4 million on Pen Windows was not to build a better mousetrap but to destroy GO's revenue by spreading fear, uncertainty, and doubt among GO's potential competitors? Is it really the case that GO would have had a bright future--and that consumers would have been able to buy acceptably-performing handheld pen computers five years earlier--if not for Microsoft's use of Pen Windows as a monopoly maintenance tool? Lead Pen Windows engineer Martin Eller certainly thinks so: "Pen Windows was a winner. We shut down [the] GO [Corporation]. They spent $75 million pumping up this market, we spent $4 million shutting them down. They're toast. That company is dead. We did our job." Who better to judge?
Note that this is not "normal" competition, the kind we economists like. In "normal" competition you bankrupt your competitors by building a better or a cheaper mousetrap. In this kind of competition you bankrupt your competitors, and then there are no products left in their market segment, and so customers have to go and buy from you in the neighboring market segment that you monopolize.
This is, as the fish said to the Cat in the Hat, "not a good game."
Posted by DeLong at November 04, 2002 09:03 AM | TrackbackBut this was a result of consumer choice. Not enough people bought GO products to keep it in business. It's not like Microsoft spent the $4 million to hire a bunch of thugs and and pillage GO headquarters. The worst one can say about this incident is that MS did a darn good job bluffing. Remember, Bill Gates was (is?) and avid poker player.
Posted by: Sean Hackbarth on November 4, 2002 11:32 AMWhat am I missing here? Where's the anti-competitive behavior, or harm to consumers, in this story?
>>> Pen Windows was a dog that nobody seemed to want to buy and Pen Windows's major competitor, GO, had gone bankrupt in large part as a result of its customers' decisions not to buy the (well thought-of) GO but to wait until Microsoft's Pen Windows came out before deciding on what hand-helds to buy, if any.<<<
So consumers decide not to buy a product until they get a chance to compare it against the potential competition. After they do get the chance to compare, they don't buy either product, and both products fail.
Therefore the maker or product B has acted in an anti-competitive manner that destroyed product A and reduced consumer welfare???
Or was it: Consumers decide not to buy a product until they get a chance to compare it against the potential competition. The producer of it hasn't planned for this rather elemental possibility, so it runs out of cash.
Therefore the maker of the product has been victimized by anti-competitive behavior that has harmed consumer welfare???
>>> Is it really the case that GO would have had a bright future -- and that consumers would have been able to buy acceptably-performing handheld pen computers five years earlier -- if not for Microsoft's use of Pen Windows as a monopoly maintenance tool? <<<
I sure don't see anything that indicates the answer is "yes".
If so, then *after* Pen Windows proved to be "a dog" how come the supposedly superior technology of GO -- *visibly* superior now, to everyone -- had no value? Why couldn't GO revive the product or sell the technology to a third party? And with this technology sitting around and available, why did the world have to wait another five years for Palm to apparently reinvent the wheel?
>>> "Pen Windows was a winner. We shut down [the] GO [Corporation]. They spent $75 million pumping up this market, we spent $4 million shutting them down. They're toast. That company is dead. <<<
Nope, Pen Windows was a loser, "a dog". And this quote from one of its makers sounds to me a lot like the ego-saving standard fare one gets from members of losing teams who look around for some rationalization for believing that in some way they "really" aren't losers. (We long-time fans of the NY Jets have received a regular diet of this for about 33 years now.)
>>> Who better to judge? ... <<<
Who better than an engineer to judge business strategy and marketing plans? ... Um, no comment. ;-)
>> Note that this is not "normal" competition, the kind we economists like.<<
I dare say it is very normal competitive behavior for established firms to use the advantages of incumbency to force new market entrants into running start-up losses. And failure to budget for such losses is one of the major causes for start-up failures. Note that this in no way implies consumers are deprived by the established firms of the chance to try the start-up firm's product.
So consumers here decided to see how the market developed before committing to GO. Where is the anti-competitive behavior by MS in that?
And where is the evidence that GO's product was in fact was any better than Pen Windows? Consumers sure didn't attest to that. If it was, why couldn't GO sell it, or at least its technology, when its superiority became evident after PW proved such a dog?
The fact that the world had to wait another five years for Palm -- think how far computer technology advances in five years! -- and that Palm then readily succeeded in spite of the same handicaps GO faced (if not worse) may well instead rather indicate that the technology for the product just didn't exist at the time for anybody. (Remember the Apple Newton?)
Whatever else, from a social perspective there is no doubt that $75m + $4m worth of resources with alternative uses were wasted.
Posted by: Bob Briant on November 4, 2002 12:48 PMDo I understand?
Microsoft can limit software advances and potential competition by simply entering a field of likely advance. After all, Microsoft can just attach extra software to Windows. Attach all the sorts of functions of perhaps Norton Utilities, and who needs Norton Utilities.
Posted by: on November 4, 2002 01:12 PMThere's a very good chance that the quote from Marlin Eller is bogus; A creation of the book author to push a point of view. More likely he was telling his coworker that MS had spent less money than GO to create a product SUPERIOR GO's and if it didn't succeed it's because there was no market for pen computing in the first place.
The market for pen computing probably depends on a bunch of other technologies to be viable that are only just now becoming avilable: wireless, cheap touch-sensitive flat panel displays, etc..
I've seen people who are pushing that "MS is an evil Monopoly" line lie routinely about what MS people actually say. Somehow they seem justified in making things up because they are convinced of their rightness in battling evil. Go figure.
Hell, at the time Pen Windows cam out. Windows wasn't even a monopoly. OS/2 was still viable, and could have been a real threat if IBM had done a better job of making the upgrade path for people smoother sooner.
Posted by: John on November 4, 2002 01:54 PMRe:
>>There's a very good chance that the quote from Marlin Eller is bogus; A creation of the book author to push a point of view.<<
You are not lying--quite. The quote from Marlin Eller is a creation of the book author to push his point of view.
The author of the book is Marlin Eller.
Brad DeLong
Posted by: Brad DeLong on November 4, 2002 02:09 PMIs the competition ever exactly the kind that economists like? I imagine all competition is conducted without perfect knowledge of what the end game is. If MS had created a Pen computing system that took off, I am sure they would have considered that a success as well. So a post mortem may vary decidedly from the project's orriginal goal.
Especially in computers and software (or even more generically Information) the Network effect is enormous, and, I think, under-appreciated. The truth of the matter is that consumers are much, much better off with a single operating system, and at the same time that single operating system would be much better if it faced competition. These two forces are at odds with each other and it is unclear, to me at least, what the optimal role goverment has in this period of rapid improvement in capabilities.
That said, it seems clear that the govt should have had quicker authority for precautionary measures at a much earlier stage - perhaps requiring MS to have a special advisor on competition matters, without having to go through the binary ruling of monopolist or not.
>The truth of the matter is that consumers are much, much better off with a single operating system, and at the same time that single operating system would be much better if it faced competition. These two forces are at odds with each other and it is unclear, to me at least, what the optimal role goverment has in this period of rapid improvement in capabilities.
The operating system might be a "natural monopoly" but it does not follow that compatible applications software need also be part of that monopoly
Posted by: Bob Briant on November 4, 2002 03:16 PMThings will get really interesting once the private antitrust suits, like this,
go to trial.
There is no natural operating system monopoly. It is the interface between the operating system and the applications that is a natural for standardization.
Some ignorant people have argued that since there is next to no marginal cost for software that such things are natural monopolies. This ignores that products with competition change in response, and thus are not the same as products without.
That the response to the suffocating effects of Microsoft's behavior on innovation is reply to with "where is the harm to the consumer" misses the point. How can we measure the harm to the consumer of the ideas that were not pursued? How do we measure something that never happened?
Posted by: chris bond on November 4, 2002 04:35 PMI think there are some interesting issues here. First, this strategy by MS is very similar to their “vaporware” strategy, except here they are actually producing something (albeit poorly), resulting in a possibly greater economic inefficiency.
While I mostly agree with Brad, I’m struck by some of the immediate counter arguments that others have made. If GO was truly superior, after MS released PEN, why didn’t consumers go back to GO? (Sorry) Was GO unable to finance the time till PEN was released, at which point people would choose their obviously superior product? Now maybe we have something...MS promises a superior product in the near future...drags it’s heals...which displaces some spending, as people wait for the MS release. But remember these people have to be willing to wait. Then GO goes bankrupt, and MS finally releases its crap, or it vanishes. Brad calls this bad competition. That makes sense in a one shot game, but MS continues to do this...so why do people continue to wait for the release of products that are never released or are crap?
I think the potential network externalities matter. People only want to adopt something they know others will adopt, and if they tend to be risk averse when it comes to buying a new technology from a relatively new company versus waiting for MS, they wait. However you can’t keep a good idea done, well you can but not forever...
Jim Glass is partly right. It's both true that MS used monopoly power to kill a competitor, and Go failed to counter MS's predictable behavior. It's not clear what GO could have done, however. And the lesson for investors was clear: fund a company that could be a threat to MS and watch your investment disappear.
What's the winning strategy for GO? If they had held on, MS would have been able to give Pen Windows away to bleed Go's margin.
Posted by: yodaiken on November 5, 2002 02:09 AMBob Briant writes: "Whatever else, from a social perspective there is no doubt that $75m + $4m worth of resources with alternative uses were wasted."
Well, how many people who worked on either GO or Pen Windows eventually worked at Palm Pilot? How many design features that were in GO or Pen Windows were incorporated/copied into Palm Pilot?
The only real wasted resources would be the actual production equipment for the two products...equipment that would presumably be sold for scrap, and melted down to something new. Even the factory buildings could be later used for other purposes.
It's not like this is the Superconducting Super Collide (aka, "SSC," or "The World's Most Expensive Hole in the Ground"). ;-)
P.S. And don't even get me started on chemical weapons...costs a fortune to create them, then costs another fortune to destroy them. In the meantime, they haven't done anything to fulfill their "useful societal purpose," which is...to kill people! ;-)
Posted by: Mark Bahner on November 7, 2002 09:16 AM>> While I mostly agree with Brad, I’m struck by some of the immediate counter arguments that others have made. If GO was truly superior, after MS released PEN, why didn’t consumers go back to GO?<<
Or why didn't some other firm pick up GO's supposedly superior technology? Or why didn't *Microsoft* buy or copy GO's supposedly superior technology -- as it notoriously and aggressively did in every other case where it saw a challenge to Windows?
Let's be a little analytical here, as we've been encouraged to be in the past:
The claim is that Microsoft recognized a threat to Windows in the form of GO's superior PDA technology, so it used its monopoly power to put GO out of business. Thereafter Microsoft had the choice of either ...
A) Buying/copying GO's superior technology to simultaneously build itself a new profitable line of business *and* seal off any outside threat to Windows through PDAs, by dominating that market itself; or
B) Going to sleep for a few years, ignoring the dangerous threat and new market it had identified, and then letting Palm sweep up the PDA market unobstructed.
Microsoft in fact chose B.
What's wrong with this picture? The premise, perhaps?
>>> It's both true that MS used monopoly power to kill a competitor and Go failed to counter MS's predictable behavior. It's not clear what GO could have done, however...
What's the winning strategy for GO? If they had held on, MS would have been able to give Pen Windows away to bleed Go's margin. <<<
The same way Microsoft put Palm out of business?
Two things should be made clear about the "wait until you see our product" marketing ploy used by incumbent firms versus start-ups.
#1) It has *nothing* to do with monopoly power. IBM has forever been telling purchasing managers "You'll never get fired for buying IBM, so why risk buying from a start-up? Wait for us!" And so does every other established firm facing a start-up in every industry. If there are 10 established firms in the industry, they all say it. This is just business 101.
If there had been three big competing OS makers each with 33% of the market, each would have said the same thing to GO's would-be customers --"Wait for our product!" GO's situation then would have been worse because it would've been three times as likely that one of the other products would be better than its, so customers would have had three times the reason to wait before buying. GO was actually better off for having only one competitor doing this.
#2) The ploy is *pro*-competitive, not anti-competitive, in that it gives the customer more choice. If the 10 companies in an industry are prohibited from announcing plans to match a start-up's new product before their own versions are actually coming off the assembly line, consumers will have no choice but to go with the start-up's until then, even if it is producing crap. But if the 10 can announce, "we'll be making that too soon, wait for us", the consumer will have much more choice and be able to say "OK, I'll wait and see who's best" -- and that's *competition*. If the start-up doesn't budget for that, too bad for it. The competition rules are meant to protect consumers, not start-ups.
So it was too bad for *consumers* that GO faced only one incumbent using this ploy, rather than three or ten. But it was better for GO that it faced only one.
Anti-competitive behavior takes away choice from consumers. If Microsoft had said "Anyone who uses GO products can't use Windows", that would have taken away choice and been anti-competitive. Its saying "wait and consider our product" gave consumers an extra choice and was pro competitive. For MS had no power at all to keep consumers from saying, "We've seen GO's product and we like it, why should we wait for you?" -- just like they did with Palm.
Microsoft committed some serious anti-competitive offenses after it attained its dominant position. One can see the list of them, including the ones for which it didn't even pretend to have a defense, in the Court of Appeals decision at http://ecfp.cadc.uscourts.gov/MS-Docs/1720/0.pdf
But really, sometimes it seems whenever MS breathes or blinks someone says "monopoly abuse!", and it ain't so.
Posted by: Jim Glass on November 7, 2002 12:03 PMThe market can only be said to have worked if a better product was the result at the time. It's no good saying that the benefits in terms of R+D were felt x years down the line. For one company to use the R+D that some other product, or company, paid for x years ago is one of those things that show that computer technology is, for good or ill, not a free market.
Are MS's products better products? If you really believe that a competitive free market is operating, you'd have to say "Yes, every single one of MS's best-selling applications is pretty darn good, because the consumer gets a choice between several alternatives" (sounds of loud coughing ensue).
MS reminds me of the NHS in UK - inefficient, badly-designed, but universal, cheap (in comparison to buying other applications), and with no effort required. The "choice" of getting different and objectively better applications comes with disadvantages that spring exclusively from MS's incompatibility with other systems and the secrecy of its code so that it's no choice at all for most people.
Posted by: Thomas Dent on November 8, 2002 10:21 AM>>> The market can only be said to have worked if a better product was the result at the time. It's no good saying that the benefits in terms of R+D were felt x years down the line.<<<
Well, there was nothing preventing consumers from buying GO's product at the time that wasn't also preventing consumers from buying Palm's product at the time -- when they bought it aplenty.
So what is there to indicate GO had a better product at the time?
>>> If you really believe that a competitive free market is operating, you'd have to say "Yes, every single one of MS's best-selling applications is pretty darn good, because the consumer gets a choice between several alternatives..." <<<
One might keep in mind that MS indeed obtained its dominance as the result of the competitive free market operating, in the face of many alternatives, starting from the position of a workers-sleeping-on-the-floor start-up competing against established firms hugely larger than it which had all the advantages of incumbency attributed to it now.
One could recall, for instance, how Gates wrote to Sculley of Apple asking for Apple to lower the price of its OS to increase its distribution, so MS could sell more applications for it -- and how Sculley didn't even deign to reply, as he considered MS such a minor player in his game. After which Gates adopted his recommended strategy for MS and outcompeted Apple with it, from behind.
Sculley's strategy was high prices, high margins, high profits. Gates' strategy was low prices, low margins, high volume, network effects, high profits. That was the strategy that won in the competitive market. Nothing prevented Apple or the others in the market from using it.
And, self-evidently, MS's extraordinary success in going from start-up *to* market dominance can hardly be attributed to its later monopolistic position, unless we believe in causation that works backwards through time. So, obviously, there had to be something fundamentally *right* about MS's product strategy and execution to account for such competitive success.
Britain's NHS, on the other hand, seems to have been established and maintained by political fiat, a rather different process.
"For one company to use the R+D that some other product, or company, paid for x years ago is one of those things that show that computer technology is, for good or ill, not a free market."
What is your definition of a free market then...that it always produces products that are the "best"? (I can't even imagine how you'd decide that...is a Corvette "better" than a Chevy Suburban, or a Chevette? Well, maybe a Chevette, but...)
My personal definition of a free market is that people are free to buy and sell at the prices they choose. Whether or not they buy "well" (i.e. "good" goods) doesn't enter into it.
Posted by: Mark Bahner on November 14, 2002 04:13 PM