Doing Capitalism in the Innovation Economy: Markets, Speculation and the State

• Introduction
• Remarks
• Questions

JOANNE MYERS: Good afternoon. I'm Joanne Myers, director of Public Affairs Programs, and on behalf of the Carnegie Council, I would like to thank you all for joining us.

Our speaker is William Janeway, who is here to discuss his book, which is entitled Doing Capitalism in the Innovation Economy. This book has been recognized as one of the best economic books of 2012 by both the Financial Times and by Foreign Affairs. We are delighted to have him as our guest.

If you ask Mr. Janeway to tell you about himself, he might tell you about the hybrid life he leads. It is a life that has been rooted in economics and has afforded him the opportunity to pursue two careers, one as an economic theorist, who earned his Ph.D. in economics from Cambridge, the other as a Wall Street financier, who for decades has led the Warburg Pincus technology investment team that provided backing to a series of companies making critical contributions to the Internet economy.

While there are venture capitalists who have written memoirs and academics who have published theories of economic growth, in writing this book, our guest has succeeded in bringing both of his interests to life, to make the case that for innovation and economic growth, three things are necessary: market innovators, governments, and financiers.

One could easily argue that innovation is crucial to a vibrant economy. Without it, there can be no prosperity. Countries that encourage adoption of new technologies grow faster than those that do not. In fact, since the 18th century, virtually all of the economic growth that has occurred throughout the world can ultimately be attributable to innovation, whether we are talking about railways, electricity grids, or superhighways.

Still, with all this progress, there appears to be one underlying theme, which is that, in addition to the financiers and venture capitalists supporting these projects, there is the key role that government plays in encouraging and financing innovation. Recent examples can be found by looking at gains in information technology, which can be traced back to the Pentagon's role in early research or contributions made to medicine as a result of experiments conducted at the National Institutes of Health [NIH]. Governments also help stabilize the economy when private demand fails to fully employ a country's resources. Finally, governments limit the damage to the economy caused by unavoidable episodic financial bubbles. I'm sure that sounds familiar.

In an era when some see our economy stagnating—although the Dow closed today with a gain of more than 125 points—Bill's message is a powerful one. Innovation is not only the product of a great idea, but one that requires the state to play a central role, not only as a source for funding research, but as a guarantor of financial support, especially when financial speculation ends in disaster, as it so often can. His argument is one that could jumpstart the economy.

Please join me in welcoming the one person who can help us understand the economic challenges of the state and the intertwining of private entrepreneurial innovation and financial capitalism, our speaker, Bill Janeway.

Thank you for coming.

WILLIAM JANEWAY: Thank you very much, Joanne. Good afternoon.

I'm delighted to be here and delighted to have this opportunity to share with you what I learned by doing capitalism over the long generation since I left academia, in 1970-71, until I returned, about five years ago. I'll conclude with some, I think, positive thoughts about the conditions of return.

But let me begin by telling you what I mean by the "innovation economy." I'm talking about a set of interrelated, complicated processes that interact with each other and that we can envision beginning upstream in the activities of scientific discovery and technical invention, from which economically significant innovations emerge. Downstream, I'm talking about the exploration of what those commercial opportunities actually are, how they can be instantiated in businesses that generate profit and deliver value to customers.

Both of those processes proceed by trial and error—and error and error. In between, the innovations that have transformed the architecture of the market economy over 250 years have been deployed as networks whose value in use cannot be estimated at the outset of deployment, until they are actually built and used, from the canals, the railroads, the telegraph, telephony, electricity networks, highways, and, of course, the Internet.

These three different phases through which the innovation economy works share attributes. First, there are complex feedback loops between the core technology—whether it's the steam engine, whether it's electricity, whether it's digital computing—and the application domains where they are deployed, which create demands for particular kinds of advances, at the same time as the general improvement in the core technology creates new opportunities. That's what I mean by "new economic space" for exploration.

To the dismay and frustration of many relatively unimaginative mainstream economists, the second attribute is that these processes cannot be reduced to what the economists would be able to model in formal terms—to use the jargon of the trade, as the "optimal intertemporal allocation of resources." As I will explain in some detail, waste, necessary waste, is essential and inescapable in the evolution of the innovation economy.

Finally, all of these phases—upstream discovery, deployment of new networks, exploration of the economic space thereby created—depend on sources of funding that are decoupled from a narrow and immediate concern with economic value. They are dependent, historically—and logically, it turns out—on financial speculation and the state.

So let me begin by talking about the role of the state. For all economies, all nations, playing catch-up, mercantilist policies, or protection and subsidy, direct or indirect, have been central. Sometimes I think that the easiest way to grasp what has been going on in China for a generation is to imagine Chinese leadership as attempting to compress the four generations that it took the United States to catch up with the United Kingdom into one generation, with a great deal of waste along the way.

But even before that, the most neglected economist of the 19th century was the German Friedrich List. His great book was called The National Systems of Political Economy. I can tell you how neglected he is. The copy in the New York Public Library is a translation of 1873. That's the most recent one.

List wrote in the 1850s that if Britain had practiced over the previous 250 years the doctrine of free trade and laissez-faire that it preached to the world—as it had become the leading industrial nation—to use his term, "England would remain the sheep yard for the Hanseatic League," shipping out wool and buying back cloth.

Investment at scale in new technological innovation by the government, by the state, requires politically legitimate missions:

• National development, as in subsidies to the transcontinental railroads that were the physical enabler of the fulfillment of America's manifest destiny 150 years ago.
• National security, as in the role of the Defense Department, as I will discuss in a bit more detail, in constructing the foundations and leading the charge in using the new digital technologies that emerged in the mid-20th century.
• And, of course, conquering disease. No one has ever subjected the National Institutes of Health to an audited cost-benefit analysis. It transcends that kind of economic calculus.

The role of the state in the innovation economy is not merely a matter of correcting minor tactical market failures. It has required and has reflected transcendent political missions that command legitimacy from a consensus.

It has been most visible over the last 200 years in the deployment of those networks, from when DeWitt Clinton mobilized the credit of New York state to guarantee the bonds that funded the construction of the Erie Canal, thereby rendering New York the leading commercial and financial city in North America, through the manner in which, in the 1860s and 1870s, the United States government took 9 percent of the landmass of what was not yet then the lower 48 and gave it to the corrupt promoters of the transcontinental railroad to tie California to the union and, as I said, to open up the West to America's manifest destiny.

But further on, at the local and regional and state levels, the electricity grids required guarantees of monopoly participation for the power utilities suffering, as always, from the network economies that come with these networks, the perverse economics that require massive investment up front, with marginal cost equal to zero for delivering incremental service, whether it be units of electricity or ton-miles on a railroad. This means that left to competition without government sponsorship and supervision, the competitive forces will drive prices to below the level necessary to support the debt incurred to build the network in the first place.

J.P. Morgan spent the last 30 years of his life vainly attempting to keep the entrepreneurial railroad barons, whom he had funded, from cutting each other's throats, and thereby creating losses for the bondholders to whom he had sold their debt. It took the Interstate Commerce Commission to end the railroad wars.

In any case, following from World War II, a new mission for the state was created having to do with the commitment to fund fundamental science. Prior to that, fundamental science had been funded, and it had been funded by sources that were, like the state, like the Defense Department, not concerned with a financial return on investment. Originally it was the philanthropic angels of the First Industrial Revolution. Darwin's father was an extremely successful angel investor in Birmingham in the First Industrial Revolution.

I like to tell the alumni of Cambridge, back when I was wearing my hat as co-chair of the 800th anniversary campaign, that the Cavendish Laboratory, the first physics laboratory in the English-speaking world, is not named that because the natural scientists of mid-19th-century Cambridge all got together to build a monument to the 7th Duke of Devonshire. William Cavendish, that duke, was the one who gave the money and got his name on it.

By the turn of the 20th century, with the rise of the great industrial monopolies of the Second Industrial Revolution, a new source of funding for science emerged. It was the General Electrics to the IBMs to the Xeroxes —the great monopolistic corporations, who had a choice with respect to the enormous surplus economic rents they were earning. Neoclassical economists would have told them that the rational and efficient thing to do was to return the money to their stockholders in dividends or to give it to their customers by cutting price. Instead, they funded Nobel Prize winners.

In particular, of course, AT&T, the legislated monopoly to provide telephony service across the United States, created Bell Labs and, under supervision from the U.S. Justice Department, could only use the technology itself for communications. The rest of all of the digital electronics developed at Bell Labs essentially became a reservoir available for those pioneering the nascent computer industry in the second half of the 19th century.

The U.S. Defense Department played the dominant role in constructing the digital new economy, from silicon through software to the Internet, both as a funder of the science and the conversion of science into available technology, but also, very importantly, as a creative and collaborative customer, pulling companies, small and new ventures, as well as the IBMs and the AT&Ts, down the learning curve to low-cost, reliable production. This created a platform on which, over the succeeding generation, venture capitalists and the entrepreneurs we backed could dance.

Let me turn to the other institutional source of funding for the innovation economy: speculation—wild-eyed, crazy financial manias. The first point is that financial bubbles are banal. Long before the First Industrial Revolution, whenever a liquid market in assets emerged, there you would find herding behavior, momentum investing, derivatives enabling leveraged bets on the future. Bubbles and crashes—tulip bulbs in 1630s Amsterdam, the new joint stock companies of the 1690s a generation before the South Sea bubble. Bubbles are banal.

The second point about bubbles is that the target of speculation defies the imagination. Anything can be the subject of a bubble—tulip bulbs indeed, goldmines in Nicaragua, the debt of emergent countries the economic value of which nobody has any idea whatsoever. That was what created the Crisis of 1825, which was the first time a central bank had to bail out its bankers. The Bank of England stepped in, under enormous pressure, to save the new financial system of Britain. Again and again and again, the target has been real estate, including those beach houses in the Nevada desert that we just built a ton of.

But occasionally the target of speculation—and this is the third point—is one of those fundamental technologies, which, when deployed at scale, changes everything, creates a new economy. Usually the financial speculation is premature and the beneficiaries, the participants in that new economy, are not necessarily the ones who provided critical funding. As I like to say, it takes 100 Pets.com to find one Amazon.

A way to think about this complicated historical process is, in my view, through a very simple matrix, a very simple two-by-two matrix. I don't have to draw it up. You don't need a graphic. It's very simple. Just think of one dimension, one axis, as being the object of speculation, ranging from some asset or representation of an asset, like a tulip bulb, which offers no prospect of increasing productivity or potentially increasing living standards, up through railroads, the Internet, computers—a productivity-enhancing, speculative target, whose value cannot yet be determined with any certainty, except that it's bigger than a breadbox; along the other axis, the locus of speculation. Historically, if the speculation, if the bubble, has been limited to the liquid markets in traded securities—typically common stocks, preferred stocks, and bonds—the damage when the bubble inevitably bursts is limited.

Contrast 2001 with 2009. When the bubble infects the banking system, infects the system of providing credit that provides the working capital that supports the ordinary market economy of production and consumption, of work and saving, then when it breaks, it takes the economy down with it. A very simple rule for me is, let an equity bubble run, but when the banks start lending their customers the interest to pretend that the loans are still good, come down on them like a ton of bricks.

The history of American venture capital, so often identified, not only by venture capitalists, but almost always by venture capitalists, as a miraculous instrument of innovation and as potentially a tool of state policy to encourage innovation—the actual history of American venture capital, as it has been exhaustively analyzed by a range of excellent scholars, with massive quantities of data at hand—that history exemplifies what I've just been saying. There are two basic statistical facts you need to know.

First, since 1980, since there essentially became something that you could call a venture capital industry, 80 percent of all the dollars invested have gone into just two sectors: information and communications technology, and biomedicine. I argue in my book that it is not a coincidence that those are the two sectors where the federal government for decades invested masses, tens and hundreds of billions of dollars, to construct that platform, to bring the technologies close enough to commercial entry so that the necessarily limited time horizon of venture capitalists and of the investors riding on their backs could be relevant to the commercialization process.

Across a whole spectrum of other technologies, including the canonical—everyone in this room has seen the movie The Graduate. Plastics were commercialized by General Electric and DuPont. It took 30 years and tens of billions of dollars to bring the polymer science into mass production and common, everyday use—not a place for venture capitalists to play.

The point here is that one of the two dimensions through which to understand the success of American venture capital is to begin with that focus on the sectors where the government built the platform. The second statistical set of facts is about the returns venture capitalists have earned and the extraordinarily close correlation between those returns and the state of the public equity markets.

In particular, in my own academic research, we were able to look at and to analyze on a daily basis the flow of cash between the limited partners, the venture fund, and the companies that the venture funds invested in, the flow to them and the flow back. That meant that we could take those same flows and, in effect, create a virtual fund, a virtual what would now be called an exchange-traded fund, by taking the money and putting it into the stock market index instead of into the venture fund and taking it out of the index when the money flowed back to the limited partners.

What that demonstrated is that from a sample of superior venture funds, better than the industry as a whole, the median return, the 50th percentile—where you would be if you had taken the money and done a distribution, in effect, randomly into the whole set—was exactly the same as you would have gotten by investing in the NASDAQ index. The returns were terrific through the top of the bubble. But so was the NASDAQ index, which went 1,000 to 5,000 in barely five years.

Since 2001, American venture capitalists have returned a lower rate than you would have received by investing in the NASDAQ index. The enormous increase in capital committed to venture capitalists—members of the National Venture Captial Association went from managing $25 billion in 1995 to managing $250 billion six years later—that is coming down. Venture capital is, if you like, reverting to the mean.

But its history illustrates, as I say, this dynamic between state sponsorship and financial speculation.

I would summarize what I've been saying by taking up that term that I used early in this chat—waste, taking waste seriously, the necessary waste that comes from discovery, deployment, and exploration. In a simple term, one that I love to use in academic departments of economics and finance, efficiency is the enemy of innovation.

Pursuit of efficiency not only discourages the toleration of that necessary waste on the supply side of the economy—which I would honor the great economist Joseph Schumpeter and his theory of creative destruction by calling the Schumpeterian waste—but it does something else: It also encourages the toleration of unnecessary waste on the demand side of the economy. In honor of his valiant and unsuccessful fight against that unnecessary waste in the 1930s, I call that Keynesian waste.

Keynes recognized, back as early as 1929-30, that the marginal productivity, the productivity of unemployed resources, is negative. Skills rust; machines atrophy. Any mode of state investment that generated employment and income would be better than nothing. He has a characteristically provocative paragraph in the General Theory in which he hypothesizes that if the British Treasury had taken pound notes, put them in old bottles, buried the bottles in disused coalmines, covered them up with municipal rubbish, and relied on the profit motive to dig up the bottles and spend the money, and thereby create income and employment, it would have been better than nothing.

Of course, it was provocative, but equally, of course, when full employment returned to the Western world at the end of the 1930s, it was because of the greatest investment in waste imaginable—mobilization for total war.

From Joe Stiglitz, 20 years ago, endogenous growth cycles, to Brad DeLong and Larry Summers in their very highly recognized article last summer, it is understood that persistent unemployment feeds back to reduce risky investments on the supply side of the economy, thereby reducing potential economic growth. If you like, in an economy where Keynesian waste is at a minimum, in a high-growth, high-employment economy, there will be more creation and less destruction. More innovative projects will be funded, and those who were stranded in legacy employment will be redeployed more rapidly.

Let me conclude by assessing, if you like, the current state of the innovation economy. It comes in two very distinct pieces.

On the one hand, contrary to the technological pessimists, that range from the distinguished Robert Gordon, professor at Northwestern University, to the eccentric entrepreneur Peter Thiel, resident in the heart of Silicon Valley, I think it is quite clear that the digital new economy is barely halfway through the exploration phase of learning what to do with this stuff. I'll be happy to give you some thoughts about what I think the next wave is. But that momentum is fully developed in the private sector. It does not need much in the way of incremental government support subsidy or sponsorship.

However, here's the frustration. The next new economy—we can already get a sense of it. I tend to call it the low-carbon economy, because some of it will come out of alternative energy sources. Some of it will come out of technologies that reduce carbon consumption. Some of it will come out of policies that have nothing to do with technology and have a lot to do with common sense. But that next new economy does depend on an enormous amount of science not yet completed. Battery technology is woefully slow to evolve. Alternative energy sources require the same kind of massive investment and massive deployment that silicon required 50 years ago.

In the United States, the innovation economy has stalled on the back of 35 years of rendering the state illegitimate as an economic actor. I applaud the president's efforts to get this back on the political agenda, at the same time as I have to note that, for understandable reasons, the administration made a tactical error that has had strategic cost in its mobilization of the loan guarantee program that the Bush administration had created, as an instrument of the stimulus bill. You can understand it. Anyone who has anything to do with the innovation economy and venture capital knows that the most certain way to get money spent is to give it to a startup. I promise you it will be spent. And it was spent by Solyndra and it was spent by A123 and it is being spent by Tesla.

But guaranteeing hundreds of millions of dollar in loans to wannabe winners is the exact opposite of what the Defense Department and NASA did so successfully 50 years ago. They, in effect, created competitive processes where big companies, old companies, and new companies could compete to deliver innovative products and the department played the role—agencies, the Air Force in particular with respect to microprocessors—as I say, of that collaborative customer, getting the best of both worlds, pulling into production, before it could be produced for profit, new technologies, which, having been brought down the learning curve to low-cost and reliable manufacture, were then available to the commercial market. That's what we should be doing now in this array of technologies.

As we are stalled and Europe is so fixated on the oxymoronic strategy—every time I use that word, I always think, are those first two syllables really necessary?—of expansionary fiscal austerity, China is obviously playing for the leadership in the next wave of the innovation economy.

I can't conclude on an utterly pessimistic note. The fellow who hired me at Warburg Pincus 25 years ago, John Vogelstein, always used to say that you could not survive as a venture capitalist if you were a pessimist. So I will end on an optimistic note.

The global financial crisis of 2008 and the Great Recession of 2009 and continuing have had one unequivocally positive impact, and that is on the destructively separated disciplines of economics and finance. The world has broken in, if you like. Economists and finance theorists alike have, in effect, been mugged by the markets into a recognition that their mechanical models of efficient analytical solutions actually do not correspond with the way the world works and, even better, that they really have to start talking to each other, understanding that we live in one financial economy; money is not a veil behind which the real economy does its magic; money is an asset which any one of us can choose to hold in order not to make a decision, not to make a commitment, in order to self-insure against a future that cannot be anticipated fully and cannot be hedged in any other way than by holding excess cash.

That is why it is such an intellectual delight to return to the academy some 35, 38 years after I left it and to have the opportunity to reflect on what I learned by doing capitalism from a more, if you like, theoretical perspective.

Thank you very much.

QUESTION: James Starkman. Thank you for a fascinating discussion.

How would you compare the "innovation economy" in China—the development of it over the last 25 years—with our own balance of venture capital and government sponsorship?

WILLIAM JANEWAY: As I say, I think the real comparison, the more meaningful comparison, may be not so much with the U.S. today, but with the U.S. 100 years ago or 150 years ago even.

Now, obviously, you have a very different circumstance. In the U.S., the role of the state—in fact, for the world at large, the first map of the role of an active state was constructed by Alexander Hamilton and then by Henry Clay. Their attempt to deploy it in what Henry Clay called the American system, which we remember from our American history textbooks—remember, internal improvements, turnpikes and canals?—was stalled and then blocked, first by the Jacksonian rebellion against control by the bankers and the power of finance, and then reinforced by—something the Chinese know well—the enormous pervasive corruption that surrounded the whole process.

In the United States, that led to a pullback of the state, except for two primary programs: one, protectionism for manufacturing, which can be done either by an undervalued exchange rate—which the Chinese now have much less of—or by tariffs. The United States, of course, had very high tariffs from the 1840s, particularly from the 1860s through the 1920s—only the railroads at the national level. But at the state level, there was continuing what has been called competitive state, meaning American state, mercantilism that was deployed and continues to be. That's exactly what's going on when we compete with New Jersey by giving away tax benefits for corporations to stay here. It's exactly what was going on when different states and even cities were competing with railroad companies to get the tracks laid through their territory rather than next door.

The corruption is pervasive. The corruption tax can feed back politically. But the Chinese are certainly doing a pretty good job of emulating us in the massive overbuilding of infrastructure. We built five trunk railroads between New York and Chicago to guarantee that none of them would ever make any money. The Chinese, when you look at their high-speed rail network, are at least as wasteful.

This is a sensitive subject, but like the Chinese, we began trying to catch up to the frontier of innovation with that other instrument—namely, the theft of intellectual property.

Has anybody been to the Slater Mill on the Blackstone River in Pawtucket, north of Providence? Slater was a Brit at a time in the late 18th century when the export of textile manufacturing equipment was banned. It was a felony, which meant it was punishable by death or transportation. He memorized both the layout of Arkwright's mill and the way the individual machines—the looms, the mules—were put together. He got to America, worked his way up the coast, I think from Savannah, until he got to Providence, and Moses Brown backed him in the first textile mill that made money in the United States, before Lowell and Lawrence, Massachusetts.

What you have to hope for in China is that local innovation becomes powerful enough that some minimal respect for other people's ideas becomes instantiated.

It's a very controversial subject now. And it should be. Of course, in terms of efficiency, it's a great deal more efficient in the world of the Internet than it was when it took Slater six or eight weeks to cross the Atlantic and another two or three months to work his way up and find Moses Brown.

But that is also a constant.

Whether the Chinese polity can hold together under the extraordinary stresses, which Britain faced in the 1820s, of a more or less corrupt elite trying to hold the lid on the greatest explosion of economic energy and financial wealth in the history of the human species—whether they can hold the lid on that and continue to assert authority—the Brits had the Bank of England; they have the Communist Party. They both had the threat of chaos, that the public could look and see, whether it's the Cultural Revolution or the French Revolution. Whether they can hold the lid on it and migrate into a more stable political balance—we fought the Civil War, so it's not for us to say that political stability is guaranteed through this process of extraordinarily rapid change.

But I think that reading China through the lens of past nations working to catch up and to get to the frontier is a useful exercise.

QUESTION: Allen Young.

The term "industrial policy" was a very popular term at one time, not used very much now. But your suggestion of the state getting involved and making investments and so on is exactly what industrial policy is all about. Some of it was very successful; some of it was very unsuccessful. What factors do you think made for success in industrial policy and what factors made for failure?

WILLIAM JANEWAY: I think it's very important to make a big distinction between those who are working to catch up, where the economic value of what previously was innovative has been demonstrated, and those who are at the frontier.

You look at Japan, from textiles to automobiles, at Korea post-1960, following in the digital technologies. They had a very straightforward path. They got on that road through state industrial policies, which created national champions, provided privileged access to capital and to markets. And it worked. It worked very effectively.

It's very different if you shift to those who are at the frontier. There, picking winners is clearly a mistake. It was a mistake in France and Britain and Germany in the 1960s and 1970s and 1980s, when each of them had a national champion computer company, which almost inevitably was condemned not to be able to compete in international markets.

The hope—and this is a big challenge for, explicitly, industrial policy at the frontier. Once it's defined as industrial policy, it's going to be subject to judgment as to how efficiently the money has been spent. I just spent the better part of 45 minutes explaining to you that tolerance of waste is the virtue of the innovation economy.

We have a program in Washington now, which the people responsible for implementing it are terribly embarrassed about, not just because it has the single worst acronym in the history of the English language: STAR METRICS, Science and Technology for America's Reinvestment: Measuring the Effect of Research on Innovation, Competitiveness and Science.

It was imposed by Congress on the National Science Foundation [NSF] and the NIH. It's being used retrospectively to go back and try to measure some of the gains from previous investments at the frontier of science. But, of course, when you start doing that, you find things like number theory. NSF has had a tiny little budget for number theory for something like 50 years. When they did that, they didn't do it because they knew that Larry Page and Sergey Brin were going to get an NSF grant and, by using advanced number theory, develop the page-rank algorithm.

So what are we trying to find out from this? There's a continuing pressure to try to focus, measure, monitor. Someone I knew very well, Bill Proxmire, had the Golden Fleece Awards for the NSF, which was an outrageous misuse of a public platform for very narrow political gain.

That's why I say that at the frontier it has turned out to be that these politically legitimate, transcendent measures that you don't call industrial policy—it's national security. We didn't do Project Apollo as an industrial policy to create a space industry. We did it because we had to beat the Russians.

The political economy of this I think is very interesting. When it's explicitly industrial policy, as I say, that can work when you have a very well-defined, economically proven catch-up opportunity, but not at the frontier.

QUESTION: I'm Randy Smith. I'm with Capital Counsel.

When you mentioned the Slater Mill and Arkwright, I was surprised that you didn't go into the stimulus to innovation that has come from the ability of the inventor to own patents and how that has evolved now in a way where it may be indeed strangling innovation. I would like your comments on that.

WILLIAM JANEWAY: Randy, that's an excellent question. It's one that I do, in fact, discuss in my book at quite some length.

You're absolutely right, there's a kind of evolution with respect to intellectual property protection that is complicated and not linear. It begins in the United States with that remarkable item in the Constitution about patents and copyrights. Naomi Lamoreaux and the late Ken Sokoloff, great economic historians, did extraordinary work in working out how a market in patents evolved in the United States. It began with the notion that in the United States, unlike in Europe, and especially in Britain, where patents, in effect, were institutions for establishing favored monopolies and were very corrupt and extremely expensive to obtain—if I remember correctly, it's Dickens's Little Dorrit, which gives you an extraordinary understanding of the corruption: it's the Office of Circumlocution, whose job it is to say no—

In the United States, by 1840, patents per capita were greater than they were in the UK. There was actually a market in patents. There were patent agents who went out on the railroads and went from town to town looking for inventors and traded, wrote a contract whereby they would get a portion of the returns, they would do the patent, and then they would be able to find someone to exploit the patent.

Through the 19th century, the intellectual property regime in the United States is definitely innovation-supported.

Now, somewhere between 1900 and 1930, particularly as innovation came to depend on larger and larger quantities of capital—beginning, of course, with the fellow whose name is on this institution—there's a wonderful passage which you can find in my book where Andrew Carnegie reflects on hiring the first scientist to work on steel processes and how everybody laughed.

"Now, 20 years later," he reflects, "everybody realizes that it wasn't a folly on my part. It was enormously important and insightful wisdom."

Well, the big companies began to require that the inventors give the corporation the patent rights. That's the first big change.

But, as I said very quickly in passing, the Defense Department established a very distinctive intellectual property regime across all of the digital technologies from silicon to software. First of all, software wasn't patentable at all until the 1990s. But hardware—microelectronics, silicon devices, computers—of course, was patentable. But the Defense Department required that if you took the government's money, you had to license any patents you received on a fair and nondiscriminatory basis, openly.

Second, if the government became a significant customer for your stuff and it mattered, you actually had to actively participate in creating a second source—again, this was not industrial policy; this was national defense —a second source as a hedge in case you failed. The government sponsored partnerships. In fact, it was this little instrument company for the oil and gas industry in Houston, Texas Instruments, that taught IBM how to manufacture silicon reliably, under U.S. Defense Department sponsorship.

Now you roll forward and, I agree with you, I think we've entered a degenerate phase, whereby the patent wars, the extraordinary accumulation of patents, began when the great monopolies—they did two things. As they lost their market position—and I'm talking about AT&T, IBM, et cetera—they did two things. They cut their central research labs. Jack Welch proclaimed that they were closing down central research at GE in pursuit of greater stockholder value. They cut central research and they started looking on their portfolios of patents as an asset to be economically exploited, not through making stuff, but through suing other people. AT&T actually began that.

So across most of the IT-related technologies, we're now in a negative-sum game. Google spent $12 billion to buy Motorola, not because it wants to make cell phones, but to protect itself against Apple and Samsung and Microsoft. That's a huge tax on the innovation process.

It's not to say that patents in some sectors are not absolutely required. You could not go into the medical device business—never—without having patents, at least to create enough value so that Johnson & Johnson will buy you. Patents are an essential tool in the innovation economy. But in the world of IT, and particularly in the world of mobile IT, it has become, in my view, a negative process. I agree with you very much.

QUESTION: I'm a student at Pace University. I have two questions.

One, with regard to economics, do you think that there is any other form of incentive possible in an innovation economy other than money? The minute you have money, you have greed as a byproduct, and then you are going to have these financial crises and all other things.

The second thing is, I would like to know your comments about the impending energy revolution in the United States. These oil companies keep talking about new technology. I have done enough on my own of research in this field, and I think it's only the ability of the oil companies to hide chemicals that they use, because of which they are able to extract oil and gas from otherwise impossible sources. Your comments on this?

WILLIAM JANEWAY: I think we can all comment on the first question.

Now I'm talking as a practitioner. My own experience with entrepreneurs has been—it appears, I think, in Chapter 4 of my book—what I call Janeway's first law of venture capital, back in the late 1980s, when I was spending a lot of time with Hy Minsky and starting to think through some of the issues that I've been discussing with you.

Janeway's first law is that all entrepreneurs lie. The ones who really get you in trouble are the ones who don't know they are. They're saying that by my efforts, I'm going to change the world. I may make a lot of money doing it. I don't think Steve Jobs was coin-operated, as they say, about good salesmen. There is an enormous entrepreneurial drive. Schumpeter writes about it eloquently.

Sometimes, I will tell you, I have wished my entrepreneurs to be more focused on money. In fact, as I say in the book, it's the job of the venture capitalist to follow the case, because the entrepreneur is going to follow her dream right off the cliff into bankruptcy and liquidation. Schumpeter has a great line in Business Cycles: "The entrepreneur qua entrepreneur loses other people's money. I was the other people."

So I have to tell you, I don't think we would abolish greed by abolishing money. That's above my pay grade as a practitioner or a theorist. It's an observation.

With respect to fracking, I am absolutely not an expert on either the technology or the regulation. It certainly seems to be the case, A, that there's an awful lot of stuff there that is going to be produced; B, that it will be produced in ways that will characterize one of the essential positive lessons from mainstream neoclassical economics, and that is, as enunciated by the great Arthur Pigou, professor of economics at Cambridge for the first half of the 20th century, in The Economics of Welfare, that there are externalities that, without strict regulation, the producer of the natural gas and the externalities will escape being charged for. This is a classic case for exactly the kind of government regulation that the most Chicago-style conservative economist qua economist can only applaud.

JOANNE MYERS: Mr. Janeway, I have to thank you for such a fascinating discussion. I think we all want to invest with you. Thank you.