6-8-99
Hi, MGV,
Yes, you are right that I could make lots of distinctions between the Valence and IBM/MSFT situations. But in a sense, that was exactly why I made my remark that I was still trying to figure out why IBM didn't buy MSFT when it (presumably) could have. I have no compelling understanding as to why sometimes the little innovators get taken out by the behemoths, and sometimes they don't, only to come back later to devour the behemoths themselves.
There are several elements to my thinking. First, there is the problem of what we should expect. While I take seriously the economists' efficient market hypothesis, I also take seriously the economists' joke about the $20 bill on the sidewalk (and also the one about why ain't we rich if we're so smart). It cautions us to not exaggerate the power or real-world relevance of our efficient market hypothesis. And it does not seem to be only a matter of information costs, or transactions costs, or the iterative process that takes time, or the fuzziness caused by different estimations of uncertainty, or whatever. There is the problem for me of reconciling the Neoclassic market efficiency hypothesis with Schumpeter's creative destruction thesis where the small innovators knock off the stodgy relics -- and in time become stodgy relics themselves. We seem to see throughout history the processes of creative destruction, and in today's world it even seems to be rapidly accelerating, yet the market efficiency hypothesis would seem to dampen the process almost to the point of ruling it out. If I had to sort all this out, I think I would look to the theories of public choice, institutions, and bureaucracy. They all suggest it is likely markets will be inefficient in ways that make possible Schumpeterian creative destruction. In the Neoclassic environment we speak of "the firm," but while a useful simplification for many purposes, the homogeneity of decision-making the term implies is really much too casual for our purposes here. When it is recognized that firms are ultimately composed of individuals, and each has his own agenda which does not necessarily align with that of the others in the firm, or the firm itself, or the shareholders, then it is easy to see the firm's behavior may not always conform to the optimum implied in standard Neoclassic theory. In the end, I think of markets not so much as being efficient (subject to all the usual stuff) as always moving toward the efficient result – except that the efficient result is itself is a moving target. Consequently, I think there is plenty of room for big mistakes, mistakes like IBM made. There is room for creative destruction.
Following this theme, the second line of my thinking is that the real world has its way of intruding on purportedly "optimal" behavior with its endless particulars. Bill was probably right when he said IBM was stupid, even if that is our 20/20 hindsight talking, but as a serious answer it really doesn't satisfy because it simply regresses the question to why IBM was stupid. Of course we know that IBM considered the PC a passing fad (and, as I recall, expected the market to total some 750,000 units by the end of the fad), but we then wonder why they did not anticipate the likelihood that Intel would produce faster chips, software writers would create applications (like Wordstar and Visicalc) of value to the public, and so forth. So was it just institutionalized "stupidity," a form of institutional failure, a case of the theory of bureaucracy in action? A not-invented-here complex? Old mainframe guys protecting turf? The particulars can go on forever. I think it is awfully hard, if not futile, to pin down why IBM was stupid, but it was, and the rest is history.
So why hasn't Valence been bought out if its technology is so damn good? Fair question, but I have no idea. As you suggest, it may well be the Valence technology isn't all that it's cranked up to be, but then again, all those particulars that characterize every potential buyer are no doubt relevant too. Maybe they are all just stupid in the IBM way; maybe they are all subject to the usual workings of the malincentives of the bureaucratic principal-agent setting. Maybe the executive suites have never heard of Valence; maybe they figure they can crush it when the time comes. Maybe the executive suites just haven't been listening to their technical staffs about the patents and don't see the problem coming – or maybe they do listen to their technical staffs and don't think there's a problem. Or maybe their legal staffs aren't worried about the patents, or maybe the legal staffs are worried but can't get the executive suites' attention. Or maybe one or more are just waiting for the right moment – waiting, say, for when bankruptcy hits and they can buy it for a song, or waiting for it to prove its product and it's a hellova bargain even if they have to pay twice as much for it. Maybe they are worrying that if they move now, they'll engender competition which will wind up costing them too much for an unproven commodity. Maybe they figure Carl and Lev and a few others control too much stock to make it cheap to buy, and their own shareholders would have a conniption if they paid what Carl and company would want. Maybe there's some sort of poison pill in the Valence by-laws – anybody know? That actually interests me.
So I have no idea why there hasn't been a race to buy Valence – hell, I don't even know there hasn't been a race! But I do know one thing: One can infer anything in the world out of all this, meaning nothing at all can be inferred from it. As I said, I'm still trying to figure out IBM and Microsoft.
Regards, lws
PS to Bill: I just can't seem to help myself! But it's back to my own work and the pool -- time's awasting and the summer's awasting...
|
