QWERTY Trilogy

The Scottish economist Neil Kay is still diligent, and has a third paper in the works. Which adds to his previous work on the claims of path dependence put forward by Brian Arthur and Paul David. The first was; Rerun the tape of history and QWERTY always wins, and followed by The QWERTY Problem. About both of which we've earlier written.

This latest provides some fascinating information about what we could call Christopher Latham Scholes (the typewriter's inventor) intellectual property game-theory'smanship. How he used a combination of patents and trade secrets to protect the key to the success of his invention. Which, in a nutshell, was what Neil Kay calls the infrequency principle. I.e., he placed infrequent combinations of letters (XDCFV, for instance) next to each other on the typebasket--which connected to the keyboard--of his machine. Since those letters would only rarely be struck in sequence, jamming of the machine was radically limited.

Which brings us back to the essence of Paul David and Brian Arthur's theorizing--that the Scholes QWERTY keyboard was merely a historical accident. Kay's papers show clearly that that was not the case at all. It was a conscious design, worked out over years. As he summarizes in his concluding remarks;

QWERTY was a consequence of creative design rather than an accident of history, indeed QWERTY was as near-optimal in terms of the crucial performance criterion of format/device compatibility as could be reasonably expected with the state of technical knowledge that existed in Sholes' time. This in turn helps reinforce arguments in [Kay's earlier papers] that QWERTY's role as "paradigm case" of inferior standard in the path dependence literature is not consistent with the evidence, and that instead the case is more consistent with path creation than with path dependence.

Which is going to be hard to swallow for this guy;

[Brian] Arthur: ....When you have increasing returns, at the outset markets are unstable and lurch back and forth according to different small events, and then lock in to one of many possible outcomes. What locks in is a function of what happened in history. The outcome in increasing returns markets depends on small events along the way. The shorthand for that is "path dependence." Meaning that small events along the way decide the outcome.

Not to mention Paul Krugman and his wife Robin Wells, who wrote in their textbook; “QWERTY problem: an inferior industry standard that has prevailed possibly because of historical accident”

VSOP

No, not a label on cognac, but maybe one that should be affixed to a glass house from which the first economist without sin is allowed to cast stones.

So how is it that economists look so bad? The answer is that too many prominent economists chose, for one reason or another, to reject the existing model. Maybe they were just trying to score points by being different; maybe they were sucked in by the approbation of the VSPs, the rewards that came from telling important people what they wanted to hear. 

But, Very Serious Other People not only make errors telling important people what they wanted to hear (about purported market failures) that have had serious negative consequences, but don't exactly respond graciously when corrected.  In this case, the very same Paul Krugman who has been lecturing (hectoring?) other economists on his blog for several days.  Back to the future;

We  have  all  seen the way that  a good  story  that  happens  not to  be true can  take  on  a  life  of its  own, and I  realize  that they  [Stan Liebowitz and Steve Margolis] are  frustrated with  the  way that  an overstated version  of the  QWERTY  story  has spread  despite their efforts  to  stop it.  That  frustration  does not,  however, justify  the  hectoring  and unprofessional  tone  of their  piece or the  way  it  misrepresents  what those of  us  who  take  path  dependence seriously have said.

Well here's what Krugman did say about it in his best selling Peddling Prosperity;

What  conservatives believe in,  above all, is the effectiveness of  free markets as ways  to  organize economic  activity.... But what if the  collective result of  those  free  choices is  to lock  in  a bad  result?  What if we end  up  stuck with an  inferior technology....And what  if another  country  manages,  with  a little timely  government intervention,  to  "lock  in"  an  advantage  in  some major  industry-and thereby  lock  us out? No,  the  story  of the  QWERTY  keyboard  is  not just  a cute  piece of  trivia ... it is  a parable that opens  our eyes  to a whole different way of thinking  about economics  It  asserts that the  outcome  of ....market  competition  often depends crucially  on  historical  accident....And  this  conclusion  is  fraught  with political  implications,  because  a sophisticated government may  try to make sure that the accidents of history  run  the way  it  wants.

Well, what if the path dependencists favorite example of being stuck with an inferior technology--the QWERTY keyboard--turns out to be a false story?  And what if they also can't provide any other examples, when challenged to do so, for their claims that there are plenty of 'QWERTY worlds' out there?

One thing they don't do is admit error.  We know that because, when QWERTY was shown not to be inferior--i.e. one couldn't retrain on the Dvorak Simplified Keyboard for two weeks and increase one's typing speed by between 20-40%, as asserted by Paul David, Brian Arthur et al--they didn't admit they were wrong.  And they still won't.

Even after almost three decades of evidence has piled up, and continues to do so, as a new paper by the Scotsman Neil Kay, Rerun the tape of history and QWERTY always wins (sorry, but it'll cost $36 to read) shows.  Kay makes innovative uses of probability theory to show just how well designed Christopher Scholes invention actually was.  And, contrary to its critics, not because it slowed typists down.

Addendum:  Neil Kay, in his answer to three comments on his paper shows that this not ancient history, by quoting from Krugman and Wells textbook, which several thousands of economics students read;

 “Government can play a useful role both in helping an industry establish a standard and helping it avoid getting trapped in an inferior standard known as the QWERTY problem” (p.536) and they define this in their glossary as;

“QWERTY problem: an inferior industry standard that has prevailed possibly because of historical accident” (p. G-12)” 

Paul, Robin, Barcelona

Be there or be square, in less than two months for Neil Kay's presentation;

This paper reviews the emergence of the QWERTY standard which in turn has lent its name to what Krugman and Wells (2006) describe as the “QWERTY problem: an inferior industry standard that has prevailed possibly because of historical accident”.  QWERTY was neither inefficient nor an accident, it was engineered by Christopher Latham Sholes in 1873 to be as near-optimal as possible given the technology and user needs of his day.  To achieve this, Sholes used a simple meta-rule that is obvious once articulated but which has not been publicly recognised until recently.  Despite its general adoption as “paradigm case” in the literature on path dependence, the analysis here finds the evidence is not consistent with QWERTY being a path dependent phenomenon, nor does QWERTY provide any basis for policy prescriptions based on common interpretations of what constitutes “the QWERTY problem”.  

Too bad thousands of impressionable economics students have already read the opposite in the Krugmans' textbook. As Kay puts it;

Krugman and Wells (2006), who in the glossary of their introductory economics textbook define; “QWERTY problem: an inferior industry standard that has prevailed possibly because of historical accident” (p. G-12)”. The QWERTY problem has as its basis “in the world of QWERTY one cannot trust markets to get it right” (Krugman, 1994, p.235).  Krugman and Wells’ (2006) advise their beginning economics students; “Government can play a useful role both in helping an industry establish a standard and helping it avoid getting trapped in an inferior standard known as the QWERTY problem” (p.536) and also “in principle government intervention might be useful in moving an industry to a superior standard” (p.534) .  

The Krugmans' confidence being in inverse proportion to the evidence they offer for the actual existence of the 'QWERTY problem'.

Feelin' Groovy

Back in the waning days of the last century Dr. Brian Arthur wasn't one to shy from tooting his own horn, as in this 1998 interview in Pretext Magazine with Dominic Gates;

Gates: .... What is path dependence, and what are network effects? 

Arthur: In high tech there are three very particular mechanisms that make for increasing returns. One is up-front costs....there are often very large up-front R&D costs....

Another effect is what I call customer groove-in. Sometimes it's just called learning effects. ....the more I type on the QWERTY typewriter keyboard, the better I get at that. It's harder for me to switch over to some alternative keyboard. .... because it's hard to learn a high tech product. So the more I use Microsoft Word... the more locked in I am or grooved-in to Microsoft. 

[Then there are] network effects [where the more people use a network, the more valuable being part of the network and using its product becomes]....

Gates: And path dependence?

Arthur: ....When you have increasing returns, at the outset markets are unstable and lurch back and forth according to different small events, and then lock in to one of many possible outcomes. What locks in is a function of what happened in history. The outcome in increasing returns markets depends on small events along the way. The shorthand for that is "path dependence." Meaning that small events along the way decide the outcome. 

As we said in an earlier post, these ideas were used to launch an anti-trust case against Microsoft (and are now threatening the same fate for Google). Brian Arthur took some pride in that in this Pretext interview;

Arthur: In 1990 I published an article called "Positive Feedbacks in the Economy" in Scientific American, which was noticed by many people. One of them was Gary Reback [the lead lawyer representing several of Microsoft's corporate rivals, including Netscape and Sun]. When certain companies in Silicon Valley were filing a white paper against the acquisition of Intuit by Microsoft--I think in the fall of 1995--Reback contacted me and asked me to take part in that. Actually, by then, I thought the case had a lot of merit. I was happy to be part of that. 

And it seems further that that early paper of mine in Scientific American, and some of my other early papers, appear to have influenced people at the Justice Department--notably Joel Klein....

All true, the misbegotten U.S. v. Microsoft was undoubtedly the result of Brian Arthur (and his Stanford colleague Paul David) misunderstanding, misapplying, or deliberately inflating the consequences of the idea of increasing returns. Which errors still bedevil the economics profession to this day.  However, in this interview Dominic Gates was a bit better prepared than Arthur had bargained for;

Gates: Let's talk about some of your critics. Stan Leibowitz [sic; 'i before e, except after c'] and Stephen Margolis have written a critique.... They claim to debunk the historical basis of path dependence theory, specifically the famous QWERTY story [that the familiar QWERTY arrangement of the keys on a typewriter was deliberately designed in the 19th century to slow typists down, because early manual typewriters tended to jam. Once typewriter manufacturers were locked into QWERTY, an alternative design that allowed faster typing failed to supplant it.] Leibowitz and Margolis call this story a "fable" and the Wall Street Journal refers to it as an "urban legend." 

Arthur became un-grooved;

Arthur: It is perfectly demonstrable that we are indeed locked into a single QWERTY keyboard. There are legions of examples of lock-in. I'm not sure even Margolis or Leibowitz would deny that. 

In fact, they most certainly do deny that, for the simple reason that there not only are not 'legions of examples', there isn't even one example from those proffered thus far (15 years after this interview was conducted) that can withstand scrutiny.  Arthur's (and David's) theory has absolutely no empirical support at all.  The interviewer presses on;

Gates: Right. But what they were saying was that it wasn't an inferior technology that locked in, and that the historical story which claimed it to be so was simply not true. 

At this point, any intellectually honest person would admit to having been wrong. Instead Arthur sets out on a voyage on that Egyptian river, Denial;

 Arthur: It's absurd to think that any theories of increasing returns hinge upon whether QWERTY is better or worse. That is nonsense.

Well, that's not exactly what is at issue--'theories of increasing returns', say Alfred Marshall's from the 19th century--is it.  What is at issue is Arthur's theory of path dependence by lock-in due to network effects.  And for that theory QWERTY is supposedly the paradigmatic example.  Not 'nonsense'.  Back to Arthur, now back peddling furiously:

If you shine the appropriate light on it, you could demonstrate that under certain circumstances something that locked in -- like QWERTY -- wasn't so bad after all. I don't know anybody who is saying QWERTY is wonderful, but it's not clear to me that QWERTY is that great. 

Who ever said anything about it being 'great'? The question is whether it meets the needs of typists as well as any other arrangement of the keyboard. It's only if it does not, that there is a better system available, but for some reason consumers can't access it--because they're locked-in to an inferior system--that the QWERTY example makes any sense in support of the theory of path dependence.  Arthur and David have been peddling a story of themselves as revolutionaries in economics, not doddering old fogies. More from the Pretext interview;

Arthur: One can take anything that locks in and at the time it locks in, normally it's better; that's why people are buying it. It's more convenient, or it's out there, or it's what you run across. But the point is that there could have been something else that might have locked in that, in the long run, may well have been better.

Of course Arthur's story had been that it was historical accident--depends on small events along the way, he'd told Gates earlier--that people bought it, not because it's better. That consumers buy products that meet their needs is hardly revolutionary or even news. Skipping ahead in the interview we finally come to complete surrender by Arthur.  As the old adage had it, when you're being run out of town on a rail try to make it look as though you're leading a parade;

Gates: But isn't an important part of your contribution your pointing out that things that get locked in aren't necessarily the best? It's not just to demonstrate lock-in, but to demonstrate lock-in of something that wasn't good for consumers. 

Arthur: Well, again, you only get excited about that if you belong to the right wing of American ideology. 

This notion that the market is always wonderful and perfect is a right-wing ideological idea. ....

The theory doesn't say that what locks in has to be inferior. The theory says that it's not necessarily superior.

Of course that is not at all true of the theory.  As Arthur's colleague Paul David put it in his most famous paper, Clio and the Economics of QWERTY in the American Economic Review in 1985, marketplace competition resulted in standardization on the wrong system (emphasis in David's paper). As Arthur himself stated in his Scientific American paper, of which he is so proud; Increasing returns mechanisms can also cause economies—even successful ones such as the US and Japan—to become locked into inferior technology-development paths .... (emphasis by HSIB)

That was the whole idea--an inferior path taken--behind Arthur's Scientific American prescription;

Steering an economy with positive feedbacks so that is chooses the best of its many possible equilibrium states requires good fortune and good timing—a feel for the moments at which beneficial change from one pattern to another is most possible. Theory can help us identify these states and times. And it can guide us in applying the right amount of effort (not too little but not too much) to dislodge locked-in structures.

Ignore where Arthur thinks he's going to find someone wise and farsighted enough to know what is the best path to be taken, much less how much of what remedy is, not too little but not too much.  Why is he promoting the need for a helmsman to do so if the marketplace selects that best path, or one equally good, on its own? Finally the interviewer asks the $64 million question;

 Gates: ....Do you have a smoking gun for increasing returns? 

Arthur: I find I'm puzzled by all of this because it's a bit like debating evolution with creationists: "But if you believe in evolution, the inference is that angels must have evolved their wings, and that would upset all of theology." For me it's moot. The onus isn't on me or anyone else, to show that we're locked in to any inferior thing. The onus is on the opinion page of the Wall Street Journal and the libertarians to show that all things that we're using in the economy are not just the best they could have been at the time, but are the best that could possibly have emerged. Nobody in computer science believes that about DOS. As for the QWERTY keyboard, if Margolis and Liebowitz can prove it's the best, my hat is off to them.

As Steve Margolis recently pointed out, while he and his partner Stan never claimed it was the best keyboard, Neil Kay has apparently done just that in Rerun the tape of history and QWERTY always wins: We find that QWERTY developed a degree of format/device compatibility that was near-optimal.... 

However that is beside the point, in the above paragraph from Arthur, he has it completely backwards.  Science requires that a theory be supported with evidence, if it is to be taken seriously.  Just asserting something, and saying it's up to everyone else to prove I'm wrong, is risible. Not to mention that no one--from Adam Smith to Milton Friedman--has ever claimed, as Arthur puts it, that all things that we're using  in the economy are the best that could possibly have emerged.

Yet Arthur is happily peddling his theory still, at the Santa Fe Institute. And being allowed to influence young, unformed minds.  Is this...well...scholarly?

[Update] We forgot to mention that (as of November 2012) Brian Arthur was still in denial over his beloved theory, in this comment on the above mentioned Neil Kay's Rerun the tape of history...

David’s claim was that with QWERTY, “markets drove the industry prematurely into standardization on the wrong system” [David’s italics]. The onus is not on Paul David to show that QWERTY was inferior to one possible keyboard, Dvorak. The onus is on his critics to show that QWERTY is superior to all other possible keyboards. I calculate that there are 2,658,271,574,788, 448,768,043,625,811,014,615,890,319,638,527,999, 999,999 of these, an admittedly large number (some 10⁵⁴). So even if someone finally and convincingly proves Dvorak inferior to QWERTY, this only goes part way toward proving QWERTY’s superiority, the critics still have all the other cases before them.

How someone who had the intellectual ability to earn a doctorate, can misread the logic of his colleague's claim so completely is astonishing.

In the Church of Path Dependence

The First Commandment is not, Brevity is the Soul of Wit, but more like Never Give an Inch. As one of its high priests, Paul David, has made all too obvious over the years in his attempts to recover some dignity after having his claim of market failure in his famous 1985 paper Clio and the Economics of QWERTY demolished by Stan Liebowitz and Steve Margolis in several papers over the years.

Let's visit a very small part of one such attempt from the year 2000, as David attempts to deny he actually said (about the QWERTY typewriter keyboard) that marketplace competition resulted in standardization on the wrong system (emphasis in David's paper);

Actually, it is within the context of static general equilibrium analysis that economists developed the concept of ‘market failure’ – namely, that the Pareto optimality of allocations arrived at via atomistically competitive markets is not guaranteed except under a stringent set of convexity conditions on production and preference sets; and, further, it requires the existence of markets for all extant and contingent commodities. One may or may not accept the usefulness for pragmatic policy purposes of defining ‘market failure’ in a way that takes those conditions as a reference ideal. Analytically, however, it remains a total non sequitur to assert that the essence of path dependence – a property defined for analyses of dynamical and stochastic processes–consists in asserting propositions regarding the possibility of ‘market failure’ that were proved first in the context of purely static and deterministic models. 

Quite the contrary proposition holds: under full convexity conditions a non tatonnement general equilibrium process can be shown to converge in a strictly path dependent manner on one among the continuum of valid ‘core’ solutions which satisfy the criterion of Pareto optimality (see Fisher 1983, and David 1997b). This should be sufficient to expose the logical error of claiming that the essential difference between models of path dependence and standard neoclassical analysis must be the former's insistence on the presence of ‘market failure’. 

The reader is spared the rest of the double talk by David that follows in this paper, as the above should be enough to at least induce the suspicion that Professor David is hoping no one actually reads what he has to say. Instead, let's see what Douglas Puffert, who was Professor David's Stanford research assistant back when the Clio paper was written, had to say recently (2010) when he had the opportunity to write an article on path dependence for the Economic History Association's encyclopedia (which is, at least, unlike David's ramblings, readable);

...market failure per se has never been the primary concern of proponents of the importance of path dependence. Even when proponents have highlighted inefficiency as one possible consequence of path dependence, this inefficiency is often the result of imperfect foresight rather than of market failure. Market failure is, however, the primary concern of Liebowitz and Margolis. This difference in perspective is one reason that the arguments of proponents and opponents have often failed to meet head on....

Market failure?  Perish the thought.

Unfortunately for both David and Puffert, Brian Arthur (the other high priest of Path Dependence) is still around to contradict them, as he did just last year in his comments on Neil Kay's Rerun the tape of history and QWERTY always wins;

QWERTY, as a standard—or better as an example of what the market has served us up in the long evolution of one particular technology—has become in economics a focal point, a rallying point for a larger issue: whether the market can lock us into an inferior standard. And this itself is part of a still larger issue: whether the free markets of capitalist economies can drive us into inferior outcomes.

..... The correct question is whether economic markets can lock in to inferior outcomes.

Schism?