Costs and Causes of Innovation
Purchase Book
The Idea Factory : Bell Labs and the Great Age of American Innovation
author: Jon Gertner
publisher: Penguin
pub date: 05.01.2013
pp: 432

Raphael Calel on The Idea Factory : Bell Labs and the Great Age of American Innovation

Costs and Causes of Innovation

November 13th, 2013 reset - +

SO ALIEN TO US would the modern world be without a telephone network, satellite communications, and computers that we rarely find examples of such imaginings outside the fantasy section in bookshops. That wondrous world is one stripped of the technologies developed at the Bell Telephone Laboratories. Indeed, if one took Arthur C. Clarke’s third law to heart — any sufficiently advanced technology is indistinguishable from magic — the scientists at Bell Labs must have appeared to be sorcerers summoning the era of advanced electronics.

This, I hope, establishes Bell Labs as a worthy subject. It does not immediately tell us, though, what might be learned from a closer study of its history. We might, for instance, learn about the people involved and the technologies they developed, which is perhaps reward enough. But apart from historical eccentricities, what insights might Bell Labs offer for current debates about the role of innovation in society?

Jon Gertner’s popular history of Bell Labs, The Idea Factory, promises such observations about the underlying causes of innovation. Indeed, the book’s cover prominently displays the following excerpt from Walter Isaacson’s review: “The Idea Factory explores one of the most critical issues of our time: What causes innovation? ” I do not wish to single out Mr. Isaacson, who is hardly responsible for the cover design, but this quotation offers an instructive starting point for considering what the book does and does not teach us about innovation.

Before getting into this, though, a few paragraphs of background material, in the interest of providing proper context for my subsequent remarks. Alexander Graham Bell, one of the inventors of the telephone, helped found the Bell Telephone Company in 1877. A series of legal maneuvers ultimately split that company into AT&T, a service provider, and Western Electric, supplier of all the equipment necessary to build and maintain the phone system. Almost from the start the US government bestowed on these two companies the privilege of operating monopolies in their respective domains. Indeed, the technology of the early telephone system had the properties of a natural monopoly, so that preventing competition might encourage the expansion and improvement of the network while avoiding costly duplication of infrastructure and standards. With their core revenue streams thus protected from day-to-day competition, the leadership was free to more fully consider the long-term trajectory of their businesses. How would, or should, the phone system look one, two, or three decades hence?

At this point, it is important that we appreciate the crudeness of the very early phone system. Phone signals were of very poor sound quality, and calls could only be transmitted over relatively short distances before speech became unintelligible. Subscriptions to phone services were expensive and only available to few. Because several households would share the same phone line, only a single conversation could be carried out at any one time (and neighbors could easily listen in). There was no dial tone or ringer, so the caller would have to yell loudly across the open phone line hoping to alert the intended recipient. It was against the backdrop of such a phone system that the executives at AT&T and Western Electric began asking what the phone system of the future should look like. The now legendary Bell Labs — famous as the antecedent of Silicon Valley, the home to seven Nobel Prizes, the birthplace of the transistor and many other inventions integral to modern electronics — was established to answer this question. Its job would be to create a vision for the future of communications, and to develop the technology needed to realize that vision.

In The Idea Factory, Gertner weaves a narrative around the life histories of a handful of the Labs’ most influential characters — primarily Mervin Kelly, Jim Fisk, William Shockley, Claude Shannon, John Pierce, and William Baker — a narrative that encourages us to contemplate profound questions about innovation. A common contention, highlighted by Mr. Isaacson’s quote earlier, is that The Idea Factory addresses the critical question of what causes innovation. By understanding the backgrounds of these men, how they all came to Bell Labs, and how they worked and innovated together there, we can peer into the clockwork of innovation and demystify its otherwise hidden mechanisms. I believe this is neither a fair nor instructive way to read the book, and may in fact blind us to the very real and valuable lessons that can be gleaned from it.

But suppose we read The Idea Factory with these blinders on: what would we learn about innovation? Was Bell Labs more successful than we might expect it to have been? Suppose that a handful of world-class laboratories are competing for the next major technological breakthrough, and that it is impossible to predict which lab will be responsible for the next discovery. Even without complicating factors like technological path dependence, chance alone would likely produce a run of successes at one of these laboratories. In other words, there may be no need to invoke a deeper set of causes for this or that laboratory’s success, and in fact, searching for patterns may only result in spurious inferences. Is the success of Bell Labs really an anomaly that deserves a special explanation?

A definitive answer is not possible, of course, but it is enlightening to consider the evidence offered in the book. Even in the biased sample of major technological inventions recorded in The Idea Factory,[1] a decided minority can be said to have originated at Bell Labs.[2] The laboratories at NASA and MIT, for instance, appear no less successful. This is not to disparage Bell Labs, just to say that its success may be overstated, and may have been less due to the Labs’ unique features than due to historical accident. Similar studies might be made of NASA or MIT, and some different unique features would be listed as causal. In other words, the success of Bell Labs or any other research facility is not sufficient proof that its example might teach us about the causes of innovation.

For the sake of argument, though, let us suppose that Bell Labs’ success is something that requires a special explanation. In this case, a careful study of Bell Labs’ history might well reveal what accounts for its rare ability to innovate, at least in principle. In practice, then, what can the history of Bell Labs, as recounted in The Idea Factory, teach us about the causes of innovation?

Gertner chronicles the youths of the great innovators and administrators at Bell Labs, how they all eventually came to the Labs, and how they worked and innovated together. There are important similarities between the innovators in their early years: they were incurably curious children all, the sort that will disassemble any mechanical device within reach just to see how it works, encouraged by some parent or devoted teacher, ultimately winning scholarships to top universities. Their recruitment to Bell Labs, according to numerous anecdotes in the book, could often be attributed to the attraction of high salaries. The success of their research, Gertner’s account suggests, derived mainly from the constant interaction of these bright and creative people in an environment of relative academic freedom.

The lessons, if one could call them that, are that: (1) some curious children, if properly encouraged and nurtured, grow into creative adults, (2) high salaries are likely to attract some of these creative people, and (3) allowing them interact and pursue their ideas with relatively little interference increases the chances that they innovate. I very much doubt that this is news to anyone. We continually reform our school systems in the hope that we will become better at spotting and promoting curiosity and talent. The high salaries paid by financial institutions and software companies undoubtedly explain why they recruit many students from top universities, and why our economy consequently produces a great deal of financial and software innovation. The account offered in The Idea Factory does not reveal new avenues for acting upon this knowledge to promote innovation.

Although The Idea Factory does not advance our understanding of the causes of innovation, it speaks to a far more important question: what are the costs of innovation? Our collective stock of knowledge is continually added to by technology start-ups, by large industrial laboratories, by government research departments, by universities, and by independently tinkering individuals. These are different “models of innovation,” if you like, and the critical policy question is how to best balance our support for these different models. This is where Gertner’s account is most successful. Bell Labs is a prominent historical exemplar of the monopolistic private research laboratory. What does The Idea Factory teach us about the benefits and costs of the Bell Labs model?

The benefits are comparatively easy to see, the list of technologies associated with the Labs well known. Bell Labs was most often the developer of new technologies, while publicly funded universities and research institutes did the basic research. This split between the public and private sectors does not appear to be peculiar to Bell, but characteristic of large private research laboratories. To give just one example, a similar division of labor has been documented in pharmaceutical innovation (see The $800 Million Pill by Merrill Goozner). The main benefit of the Bell Labs model appears to be its ability to recognize the potential value of new scientific and engineering breakthroughs, and then to develop and disseminate the resulting technologies.

But there are costs, too, of operating the Bell Labs model. Research at Bell Labs depended crucially on the augmented stream of revenues that resulted from the US government’s decision to insulate AT&T and Western Electric from direct competition. Even if monopolization is justified on technological grounds, monopoly status could allow a company to charge customers in excess of what might be needed to fund research. On top of this, allowing the monopoly to patent their innovations — conferring a monopoly with regards to intellectual property — adds an extra layer of insulation that can be used to extract further rents.

The Idea Factory demonstrates that these are not merely theoretical costs of the Bell Labs model. To protect their most valuable asset, their monopoly status, Bell ingratiated themselves with the Department of Defense. For many decades after technological advances had undermined the original technological justification for the monopoly, this relationship protected Bell from the Department of Justice’s efforts to break up the monopoly and allow competition. Repeated investigations by the Department of Justice had shown that the telephone company was abusing its monopoly power to charge excessively for their services. Years later, in a settlement with the US government, the telephone company managed to retain its protected status by agreeing to license all of their patents for free or for nominal sums to US companies, feeling it was a minor concession. This was both a display of their influence in government and highly suggestive evidence that the extra fees earned from having the second layer of monopoly protection — intellectual property — had been a windfall all along. This protection seems to have been irrelevant for supporting or encouraging innovation at Bell Labs, yet it earned Bell a great deal of money over the years at the expense of those wishing to use the technology.

Another potential cost of the Bell Labs model is that, if the Labs failed to recognize a technological opportunity, others could be effectively prevented from making that opportunity a reality. The combination of the telephone monopoly and intellectual property rights served as an impossibly high barrier to entry to potential competitors. Indeed, as soon as the telephone company began facing competition in its peripheral activities, evidence quickly emerged that the Labs sometimes did fail to recognize technological opportunities. They invested heavily in an ultimately doomed alternative to fiber optic cables, and completely overlooked the opportunity and value of integrated circuits. Perhaps there are many earlier such examples that we will never know about, and for which the foregone benefits are incalculable.

Monopolistic profits and missed opportunities were not the only costs, however. Bell reaped large financial rewards from its monopoly status, and was also allowed to become one of the few authorities on communication technology. Both assets were leveraged to gain influence in government and to protect Bell’s future revenue streams. One effect of this influence was the prolonged existence of the monopoly, and hence the multiplication of the costs discussed above. Another effect of Bell’s influence was to help open the door for corporate influence in government more generally, the continuation of which is visible in government to this day. And finally, The Idea Factory contains numerous anecdotes that make plain Bell Labs’ role in advocating and facilitating the growth of the Cold War surveillance state, a trend that doubtlessly benefited the monopolist. The cold warriors they supplied to various government agencies and commissions appear to have been honest and ideological, but it cannot be denied that their arguments benefited their employer.

In the plus column of the balance sheet, then, the Bell Labs model has produced some remarkable innovations. Its role was often not as originator of new technologies, but as one that recognized their potential and undertook the investment and research needed to create usable and mass-producible products. On the other hand, it has not been persuasively demonstrated that Bell Labs was responsible for a rate of technological progress otherwise improbable. On the contrary, there is evidence — independent discoveries and missed opportunities — that we would not have been entirely without these advances. We need to distinguish the fact that Bell Labs played an integral role in the development of modern communications and electronics from the idea that they might be irreplaceable. Perhaps, then, it is not too difficult to imagine that in a counterfactual history without Bell Labs we might still have enjoyed many of the same benefits from small competing phone companies, or physics and chemistry and engineering departments at universities, with similar levels of public investment. But while we would probably have enjoyed these benefits anyway, it seems likely that a world without Bell Labs might have spared us from excessive telephone charges and high barriers to entry for competing technologies.

It is less obvious where to add Bell’s effect on the growth of the Cold War surveillance state on our balance sheet, but here too, it is difficult to imagine that anything other than a large company, with a virtual monopoly on expertise, could have exerted such a powerful impetus toward the modern surveillance state. Without Bell Labs, its history and scope would likely have been very different.

All of this makes for a rather perplexing balance sheet. It is hard to attribute the benefits of Bell Labs to its unique features, and difficult to defend the proposition that other models of innovation would have imposed similar costs. Of course, my interpretation of the evidence might overstate or understate the social benefits and costs of Bell Labs, and the balance sheet I’ve outlined here is undoubtedly incomplete. Indeed it is, but much in the same way that the The Idea Factory provides an incomplete history of Bell Labs.

The real problem we face is to decide how our society should expend its efforts to encourage innovation. Bell Labs represents one answer; universities, government agencies, and small private companies offer different ones. On its own, The Idea Factory cannot tell us whether Bell Labs provides a good or bad model for promoting innovation — it needs to have its benefits and costs stacked up against other models. Only through debate and comparison with alternative models of innovation will we ever understand the advantages and disadvantages of each model, and only then can we hope to select policies that strike an appropriate balance among them. Gertner prepares the way for us by describing the benefits and costs of a monopolistic research lab with texture and substance. If The Idea Factory is read as a book about the causes of innovation, however, we run the risk of passing over this debate. It is inappropriate and unhelpful to read The Idea Factory with a view to learning about the causes of innovation, but I am glad that there are much better reasons to read it.


[1] A rough count would at least include vacuum tube amplifiers, radar, transistors, TWT tube amplifiers, satellites, pulse-code modulation, silicon solar cells, maser, laser, fiber optic cables, integrated circuits, the Unix operating system, CCD, and cellular telephony.

[2] Transistors, silicon solar cells, the Unix operating system, and cellular telephony.

¤

Raphael Calel is a Ciriacy-Wantrup Postdoctoral Fellow at UC Berkeley.

print

Comments