Financial Bubbles and Technological Innovations

by Michael Lind, Fellow at the New America and author of Land of Promise: An Economic History of the United States

Speculators may do no harm as bubbles on a steady stream of enterprise.  But the position is serious when enterprise becomes the bubble on a whirlpool of speculation.  When the capital development of a country becomes a by-product of the activities of a casino, the job is likely to be ill-done.  John Maynard Keynes[1]

Over the last two decades, the United States has experienced two major financial bubbles.  The first was the so-called “tech bubble,” which in spite of its negative effects helped create the information and communications revolution that has transformed the U.S. and world economies.  The second bubble, the housing bubble that developed in the wake of the tech bubble, produced no such lasting economic benefits and when it burst caused enormous damage to the economy.  The proximity of these two bubbles raises some important questions about the relationship between technological innovation and finance.  Is the overinvestment characteristic of these bubbles an unfortunate but inevitable by-product of waves of technological innovation?  Can bubbles play a positive role, by providing finance for entrepreneurs and start-ups that might otherwise be unavailable?  If the policy objective is to promote technological innovation and the diffusion of new productivity-enhancing technologies, are there “good bubbles” which can be distinguished from “bad bubbles”?  Or are all bubbles bad by definition?  And if there are positive aspects to some bubbles, how should their benefits be assessed compared to the costs of “busts” and the debt deleveraging that tends to follow bubble-driven booms?

The answers to these questions have far-reaching implications for public policy.  Should policymakers try to prevent financial bubbles, even at the risk of slowing down or even retarding innovation?  If not, should the goal of policy be to moderate bubbles, or, alternately, to compensate for the harmful effects of bubbles (for example, by pursuing compensatory fiscal policy during bubble-caused downturns)?  Or should policymakers attempt to distinguish between “good” and “bad” bubbles?

In this essay, I will review arguments that bubbles are inevitably associated with waves of innovation and perhaps even indispensable for it.  I will then argue that this approach neglects the importance of sources of innovation, including early-stage commercialization, that can occur without speculative venture capital, and that result from research and development financed or undertaken by corporations, governments, banks and universities.  Rather than accept or even celebrate bubble-driven innovation, we should consider how to use these other sources of finance to decouple innovation from destructive asset bubbles.

Joseph Schumpeter on Technological Innovation and Finance

According to the neoclassical tradition of economics, which dominates academic economics departments in the English-speaking world, competitive markets tend toward equilibrium and technological innovation must be explained in terms of a mysterious, “exogenous” X Factor known as “Solow’s Residual” after the economist Robert Solow.  A much more sophisticated approach is provided by the rival tradition of “evolutionary economics,” a branch of “institutionalist economics” pioneered by the Austrian-American economist Joseph A. Schumpeter (1883-1950).

Schumpeter is best known for coining the phrase “creative destruction.”  All too often, however, creative destruction is equated with the rise and fall of firms as a result of normal market competition.  But this is not what Schumpeter meant.  He was referring to more fundamental and transformative upheavals, including but not limited to successive industrial revolutions based on transformative “general purpose technologies” like the steam engine, the internal combustion engine, and the computer.

The context for the phrase is provided by a passage in Schumpeter’s Capitalism, Socialism, and Democracy (1942):

The opening up of new markets, foreign or domestic, and the organizational development from the craft shop to such concerns as U.S. Steel illustrate the same process of industrial mutation—if I may use that biological term—that incessantly revolutionizes the economic structure from within, incessantly destroying the old one, incessantly creating a new one.  This process of Creative Destruction is the essential fact about capitalism.[2]

For Schumpeter, then, economic progress was “endogenous,” driven by a process of “industrial mutation” or “creative destruction” that “incessantly revolutionizes the economic structure from within…”   Schumpeter believed that the “long cycles in the world economy”— waves of cyclical global growth identified by the Russian economist Nikolai Kondratiev in the 1920s—could be explained by the fact that epochal technological and organizational innovations were clustered in time (it was Schumpeter who named these cycles “Kondratiev waves” in honor of their discoverer).

In Schumpeter’s theory of economic development, the key figure was the entrepreneur (an individual or, in his later work, an institution charged with routinized innovation like a corporate research laboratory or government laboratory).  In an economy characterized by steady-state growth, lenders like banks and investors in equities act on the basis of projections of incremental growth based on existing technologies and procedures.  In Schumpeter’s words:

We will envision a society, stationary in every respect, except in that it displays a positive rate of saving….The result would, in fact, be a steady growth of the system’s industrial output by the steady addition to it of new units of plants and machinery,  which, however, must be of the same types as those which are already in use….[3]

Financing the risky, untried experiments of innovative firms requires an entirely different mindset and, perhaps, different financial institutions.  In Schumpeter’s time, this “abnormal credit” could be provided by investment bankers or universal banks, like German banks and U.S. banks before New Deal reforms, which were shareholders with representatives on the boards of the corporations in which the banks held equity stakes.  Today this function is attributed to “venture capital,” which defined broadly can include government agencies which invest in new technologies and even particular firms.[4]

Jan Kregel provides a good summary of Schumpeter’s view of the interaction of finance and technological innovation:

There is a key difference in Schumpeter’s view of the role played by money in the liberation of resources to be invested in new technology from that of his predecessors and contemporaries and is very close to the approach taken by Keynes.  The traditional view is that investment is financed by prior saving of real income, that finance is simply real resources in monetary form that is free capital to be converted into new and more productive uses.  But Schumpeter was working in the tradition of Wicksell, who recognized that in modern financial systems of the time, banks could create credit quite independently of the decisions to save out of income.  It was this approach that allowed the banks to arm entrepreneurs with the power to commandeer resources for their new, innovative projects without the need for prior saving or converting existing capital investments into a new form.  Schumpeter’s system is a Wicksellian pure-credit money system.  In Keynesian terms, the role of the financial system in the process of creative destruction is to create liquidity.[5]

For Schumpeter, then, banks and other sources of venture capital could be indispensable partners for entrepreneurial individuals and firms in the process of creative destruction.  Speculation by investment banks and universal banks could sometimes promote progress, according to Schumpeter:

“Stock exchange speculation, especially, and the speculative holding of newly issued stock were in all countries financed by banks, which, therefore, always served the purpose of financing long-term investments, at least in this indirect way, even if in no other.”[6]

Carlota Perez:  Techno-Economic Paradigms and Bubbles

The economic historian Carlota Perez, has done more than any other scholar in the Schumpeterian tradition to develop a complex theory linking long waves in the economy with transitions from one “techno-economic paradigm” to the next, with investment-driven bubbles playing a role in each historic transition. 

In a series of works Perez has developed a theory that argues that long waves can be divided into two halves— “installation” and “deployment.”  Each of these, in turn, can be divided into two periods.  The “installation” phase is divided into a period of “irruption” of new technologies followed by a period of “frenzy” or excessive investment.  Eventually the boom ends with a crash, followed by the two phases of the “deployment” era— “synergy” and “maturity.”[7] 

This recurrent pattern results from the interaction of “the spheres that co-evolve in the process of development:  technology, the economy, and social institutions.  It is the way that these changing spheres interact and influence each other that generates the sequence.”[8]

Perez has summarized these interactions in the following fashion:

• technical change occurs through clusters of innovations which form successive and distinctive technological revolutions that modernize the whole production structure;
• financial and production capital are interrelated but functionally separate agents, each pursuing profits with different criteria and behaviors; and
• the socio-institutional framework has much greater inertia and resistance to change than the techno-economic sphere, which is spurred by competitive pressures.[9]

Perez coined the term “major technology bubbles” (MTBs) “as a special class of bubbles that constitute a recurring endogenous phenomenon, caused by the way that the market economy absorbs successive technological revolutions.”[10]  She distinguishes MTBs from other bubble dynamics identified by Keynes, Charles Kindleberger, and Hyman Minsky.[11] 

Perez identifies five techno-economic paradigms in the industrial era, each of which is associated with a major technology bubble at the time of its installation.  The first industrial revolution based on “mechanization and water transport,” which she dates to 1771 in Britain, produced the “canal mania”—a bubble which collapsed in 1793.  This was followed by “the age of steam and railways” beginning in 1829 and the subsequent bust of the “railway mania” bubble in 1847.  Then came “the age of steel and heavy engineering” beginning in 1875.  During this period of industrialization, there were multiple bubbles, financed largely by the City of London as part of the “first globalization.”  As there were multiple bubbles around the world, so there were multiple crashes.  Perez dates the fourth technology paradigm, “the age of the automobile, oil and petrochemicals” to 1908, centered in the U.S.  It of course led to the 1929 Wall Street crash and the Great Depression. In her schema, the U.S. is also the center of “the age of information and digital communications” and the “Second Globalization,” whose installation phase begins in 1971 and comes to an end with the crashes of 2000 and 2007-08.[12]

According to Perez, the tech bubble of the 1990s and the asset bubble of the 2000s can be viewed as a “double bubble”: “a MTB (the 1997-2000 internet mania) followed by the easy liquidity bubble (ELB) of 2004-7.”[13]  The two kinds of bubbles have different dynamics: 

“MTBs are driven by the existence in the real economy of a clearly visible technological opportunity space promising to yield extraordinary profits and thus attracting investment money from wherever it can be found.  By contrast, ELBs are driven by the availability of cheap credit searching for whatever object of speculation is on hand or can be created by financial innovation.”[14] 

According to Perez, MTBs “are the result of opportunity pull rather than of easy credit push. But they are indeed bubbles.”[15]

In Perez’s framework, speculative finance, having served its historic purpose of funding (and over-funding) investment in new technologies and firms during the “frenzy” stage of the installment period, needs to give way to more regulated forms of finance appropriate to the succeeding deployment era, with its phases of synergy and maturity. The potential of the deployment era for producing a new Golden Age of higher productivity growth and shared prosperity may not be realized, however, if investors and firms have unrealistically high expectations of growth that represent a carry-over in attitudes from the bubble era that ended the preceding installation period:

When the boom and bust of the MTB marks the end of this installation period, most of the economy has been modernized, ample coverage of the new infrastructure is in place and new corporate giants are ready to lead the expansion by taking full advantage of the new potential.  But by this time, the financial world will have acquired the habits of being in control of investment and of getting constant high returns.  Quarterly profits will have become the main measure and production companies will find themselves forced to avoid long term projects and to constantly deliver short term gains.  For this reason, the golden age of more harmonious growth—or deployment period—that follows in the last two or three decades of each surge of development will depend on the capacity of the State to restrain the financial casino that typifies the bubble and to hand over power to production capital, allowing its longer term horizons to guide investment once more.  This usually involved changes in the financial architecture and in the incentive structure of investment.”

In the case of the tech bubble, “such regulatory and institutional changes have had to wait until the collapse of a second, much greater and more global boom and bust”—the Great Recession. According to Perez, the MTB of the late 1990s and the initial bust were followed by the inflation of an even greater ELB, instead of by stricter regulation appropriate to a new era of deployment, for several reasons.  The scandals associated with the tech bubble involved firms like “Enron, WorldCom, Tyco etc” that “were mainly related to the real economy rather than finance.” This tended to blunt any broader regulatory changes in finance. In fact, finance enjoyed an easing of regulation on capital requirements.  In addition, “constriction of demand” was absent following the tech bubble bust.  Liquidity actually increased, thanks to Chinese and other Asian monetary policies which enabled lower interest rates and credit-based demand growth in the U.S. and other western economies: 

By investing much of that [export] surplus in the advanced economies, especially in the USA, the Asian economies fed their own export markets.  The significant amount of liquidity that became available for easing credit lent more fuel to the housing bubbles that had already begun to inflate during the internet mania.  This further increased household consumption on the back of the growing value of household assets.  Hence the global imbalances that rightly can be blamed as the main causes of the bust were also an essential part of the feedback loop that generated the boom.[17]

While Perez recognizes the downsides of speculative “frenzy,” in her system the financial excesses that occur during MTBs are the growing pains of the transition from one techno-economic paradigm to the next:  “The emphasis given here to the decoupling from the real economy that occurs during bubble frenzies should not deter us from understanding the whole installation period as a time of great structural transformation.”[18]  From a great number of start-ups, a few large firms will survive to make the transition from the installation to the deployment phase: 

In general, the aftermath of the MTBs is a time of industrial restructuring, usually leading to oligopolies, in every sector of the economy.  It is the resulting new fabric of the economy, with its emerging leaders, that will carry the deployment period after recovery from the recession.[19]

Technology and Bubble Finance:  Other Views

Perez’s over-arching theory is comprehensive, elegant and plausible—but is it correct?  The work of other scholars suggests a few qualifications are in order.

In a 2011 paper entitled “Financing Risk and Bubbles of Innovation,” Ramana Nanda and Matthew Rhodes-Kropf argue “that financial activity is not purely a response to novel technologies but rather, financial markets drive innovation bubbles”—a finding that tends to blur the distinctions between Perez’s major technology bubbles (MTBs) and easy liquidity bubbles (ELBs). [20]  They agree that startups “have been associated with the initial diffusion of several technological revolutions (railway, semiconductors and computers, internet, motor cars, clean technology)” but argue that startups exploiting new technologies face two risks:  “fundamental risk (that the project gets an investment but turns out not to be viable) and financing risk (that the project needs more money to proceed but cannot get the financing even if it is fundamentally sound).”  They complain that most studies ignore financing risk: 

Thus, early round investors investing in innovative firms face an important trade-off between lowering financing risk and increasing real option value.  The most innovative firms are thus most susceptible to financing risk as they are least able to acquire a ‘war chest’ to survive a down turn. [21]

Innovative firms in an early stage of technology diffusion may benefit from “hot” markets characterized by bubbles. But they may not survive the initial frenzy.

Elsewhere, Kregel and Burlamaqui (2006) have sought to integrate the Schumpeterian framework of waves of innovation developed by Perez and others with Minsky’s theory of the three “financing positions” for assets that exist in capitalist economies:  hedge, speculative, and Ponzi finance.[22] In “Financial Experimentation, Technological Paradigm Revolutions and Financial Crises,” a contribution to a 2011 Festschritt in honor of Carlota Perez, Kregel continued to seek a synthesis of the approaches of Schumpeter/Perez and Minsky.

“Both approaches take as given the inevitability of the cyclical behavior of the economy, but also reject the idea that cycles repeat—each one is new and idiosyncratic,” Kregel writes.  “This is basically because the crucial element determining the evolution of each cycle is the technological innovations that characterize it.”[23] 

In the Information and Communications Technology  (ICT) Revolution of the 1990s, finance itself underwent a revolution that accentuated the bubble.  Kregel argues that during the installation era of ICT, the combination of deregulation of the U.S. financial system and the enabling technology of computers permitted the disintermediation of the U.S. banking system, as banks changed their role from providing direct financing for businesses or households to originating and selling financial assets that are sold and re-sold in capital markets.  According to Kregel, “the conjunction of the Perez cycle and the Minsky cycle must be sought in the concomitance of financial innovation producing new methods of liquidity creation that provide the financing for the installation of the new technology through creative destruction.”[24]  In other words, financial innovation was intimately connected with the bubble associated with the ICT revolution perhaps in ways more profound in other bubbles. 

With the bursting of the bubble, that very financial innovation would itself provoke a painful adjustment. Writing in 2011, Kregel predicted that the Great Recession would be followed by “another period of re-regulation, accompanied by a Minskyan retrenching in which financial institutions become hedge units and stop lending.”  Unless governments stepped in to bolster demand, this would “produce a global recession, if not depression, and call into question the deployment of the new technology unless the government acts to shore up aggregate demand as well as introducing new financial regulation.”[25]

Leveraged Bubbles, Major Technology Bubbles, and Easy Lending Bubbles

Another complicating factor involves the different consequences of different kinds of bubbles.  The aftermath of bubbles can be mild or severe, depending on the kind of assets involved in the bubble or the distribution of asset ownership in the population.  According to Oscar Jorda, Moritz Schularick and Alan M. Taylor:

History shows that not all bubbles are alike.  Some have enormous costs for the economy, while others blow over…When fueled by credit booms asset price bubbles increase financial crisis risks; upon collapse, they tend to be followed by deeper recessions and slower economic recoveries.  Credit-financed house price bubbles have emerged as a particularly dangerous phenomenon.[26]

This provides one explanation of why the collapse of the dot-com bubble was much less harmful in its macroeconomic effects than that of the housing mortgage bubble whose collapse nearly a decade later triggered the Great Recession.  Another explanation for the difference between the consequences of the two successive bubbles is provided by Amir Sufi and Atif Mian.

Sufi and Mian point out that the bursting of the dot.com bubble in 2000 destroyed roughly the same amount of household wealth–$6.2 trillion—as that accounted for by the real estate value lost by American households between 2007 and 2009–$$6 trillion.  The former led to “a mild recession,” during which retail spending actually increased slightly, while the housing bust “killed retail spending, which collapsed 8 percent from 2007 to 2009, one of the largest two-year drops in recorded American history.”[27] 

The difference, they argue, involves debt and who has incurred it.  In 2001 tech stocks were held almost exclusively by the rich, a very small slice of the population whose net worth allowed them to continue spending much as they had done before the crisis.  In contrast, housing was the major form of wealth for middle- and low-income Americans, who reduced their spending with disastrous macroeconomic effects because of a “negative wealth effect”—exacerbated in the case of those who could not make mortgage payments because of unemployment and lost their homes their homes as a result.

Both explanations are compatible with each other, as well as with Perez’s distinction among Major Technology Bubbles (MTBs) and Easy Lending Bubbles (ELBs).  The latter are potentially more dangerous because they produce debt widely diffused among consumers, with long-lasting harmful effects on the economy during the period of “debt deleveraging” (to use the term coined by Richard Koo).  This suggests a promising area of future research, which could include seeking to integrate historical episodes of debt deleveraging, including that of the present, into the long-wave techno-economic paradigm framework suggested by Schumpeter and developed by Perez.

Minsky on Credit, Debt and Innovation

The richest and most challenging scholarship that seeks to link Schumpeterian technological innovation with finance is found in the work of Hyman Minsky (1991-1996).  Minsky began his economics Ph.D. dissertation under Schumpeter at Harvard and completed it under Wassily Leontief, following Schumpeter’s death in 1950.  In an essay on the centenary of Schumpeter and Keynes, Minsky wrote of his ambition “to integrate Schumpeter’s vision of a resilient intertemporal capitalist process with Keynes’ hard insights into the fragility introduced into the capitalist accumulation process by some inescapable properties of financial capitalist financial structures.”[28]

According to Minsky, Schumpeter had been correct to argue in his 1912 book The Theory of Economic Development that in Schumpeter’s words “credit is essentially the creation of purchasing power for the purpose of transferring it to the entrepreneur.”[29]  According to this “endogenous” theory of credit, bankers and other lenders create credit in order to extend it to innovators whose only claim to funding is the possibility of future success; in a fractional reserve system, the entrepreneurs are not funded by stores of existing money. 

Minsky criticized Schumpeter in his later work, including Business Cycles, for moving away from this early endogenous theory of credit creation toward the “real economy” theory of money associated with Leon Walras.[30]  For Minsky, according to Mark Knell, “Endogenously created money finances the innovation process, which then generates cycles as it diffuses through the economy.”[31] Charles J. Whalen expands on this idea, observing that for Minsky:

Institutions of credit and finance are at the center of capitalist development.  While standard economic theory emphasizes exchange, Minsky’s recognition of historical time caused him to emphasize that production precedes exchange and that finance precedes production.[32]

The institutional structure of the financial sector evolved over time, thanks to financial innovation.  To the Schumpeterian idea of successive techno-economic paradigms Minsky added a historic sequence of commercial, financial, managerial and money manager capitalism.[33] 

Commercial capitalism had been associated with pre-industrial commerce.  During the early industrial era, financial capitalism developed to finance railroads and water- or steam-powered mills, providing a central role to investment bankers.  Following the Depression and World War II, the combination of oligopolistic corporations, New Deal financial reform and New Deal state capitalism reduced the role of private bankers in funding innovation.  Minsky argued that beginning in the 1980s, managerial capitalism was succeeded by money manager capitalism, in which institutional investors grew in importance.  He criticized money manager capitalism for short-termism and its contribution to volatility in the U.S. and world economies. 

In the aftermath of the Great Depression, many economists have turned to Minsky’s ideas about financial instability.[34] However, much of the commentary has tended to treat “Minsky moments” characterized by complacency and irrational exuberance as phenomena of the financial sector isolated from the historic sequence of waves of technological and economic innovation.  This ignores Minsky’s own historical approach which sought to integrate the insights of Schumpeter and Keynes into a structural explanation of qualitative differences in methods of financing.

Beyond Bubbles:  Financing Innovation without Wrecking the Economy

It is undeniable that financial bubbles often accompany waves of technological innovation.  And as Perez has pointed out, these periods of technological innovation can also result in follow-on easy liquidity bubbles.  But given the damage bubbles can do, the question we also want to ask is whether are necessary for innovation.  There are some scholars who argue they are.  Nanda and Matthews-Kropf, for example, go beyond arguing that policymakers should tolerate major technology bubbles and propose that policymakers might enable them in some circumstances: 

This implies that regulation should not always be concerned with popping bubbles, and furthermore, that those wishing to stimulate innovation should look for ways to concentrate investment in a sector or time or location in order to help create the coordination among investors that creates or magnifies innovation bubbles.[35] 

Several objections can be raised against this argument.  To begin with, Nanda and Mathews-Kropf’s argument that bubbles have beneficial effects, despite references to earlier historical episodes, is influenced largely by the precedent set by venture capital in commercializing ICT startups in the late twentieth and early twenty-first centuries. This gives a more prominent role to private venture capital in the innovation process than may be warranted. A good example of this extolling the contributions of venture capital can be found in the work of investment banker Bill Janeway, who writes: “Financial bubbles, in which returns are decoupled from economic fundamentals, are the complementary engine of Schumpeterian waste.” [36] While acknowledging the role of government-funded university research and major companies, Janeway nonetheless argues:

In defiance of Schumpeter’s expectation, economic innovation has not been effectively bureaucratized by the great corporations.  Rather, it tends to be delivered by new companies.  But funding those new companies depends on access to financiers who have access to markets prone to speculative excess.[37]

In other words, financial excess is an unavoidable part of technological innovation because such innovation is financed by venture capital, which is prone to excess.  However, this argument is flawed because it neglects other possible sources of funding for new companies or, for that matter, for innovative old companies, distinct from government funding for basic R&D.  One such source is the retained earnings of mature corporations themselves.  The history of technological innovation carried out by long-established firms like IBM and GE and AT&T does not bear out the often-repeated claim that big corporations are sclerotic and only new, young companies are innovative or that they will not finance innovation.  Even if these large established companies do not innovate themselves, many of them now follow a model of acquiring startups or even funding start-ups.  This is important because existing large corporations in theory can take a longer view than investors concerned chiefly with making money quickly with a successful IPO and therefore are less likely to engage in financial excess driven by large short-term payoffs. Thus, policy makers should pay attention to Minsky’s structural aspects of financing and, in particular, how firms use their balance sheets.

It may be true that publicly-traded companies are now under such pressure to maximize short-term returns that they are less significant sources of patient venture capital than they once were.  But there are alternate models of the corporation that insulate it from shareholder pressure.  Companies like Google, for example, have been able to invest in “moonshots” like self-driving vehicle projects in part because of near-monopoly rents and in part because of an ownership structure in which the founders retain control.  In these cases, the companies can make decisions that are largely from normal short-term Wall Street pressures.

Large firms, then, could be one source of patient capital immune to speculative frenzy and not dependent on bubbles.  Another is government, either through procurement policy or “state capitalist” loans or grants.  Many startups in many industries, from aerospace to computers, have had government—usually the military—as their first and most important customer.  Health care is another example of the central role government plays in spurring innovation.  Not only has the National Institutes of Health financed much of the research leading to new cures, but also the government is a major driving force behind the successful commercialization of new health care products.  Without major purchasers of medical goods and services through Medicare, Medicaid and the VA, much of the innovation would never reach the market.   

In addition to acting as a customer, the federal government helps innovative companies directly by means of programs like the Commerce Department’s Small Business Innovation Research (SBIR) program and the Manufacturing Extension Partnership (MEP). According to the SBIR website, the program “enables small businesses to explore their technological potential and provides the incentive to profit from its commercialization.”  Reportedly, many venture capitalists urge startups to seek funding from SBIR as well as from venture capital markets.

Other countries have similar programs for public venture capitalism.  Germany’s KfW system, for example, seeks to help young companies through every stage of growth.  Complementary services are provided by Germany’s Frauenhofer Institutes, which undertake product and process research for small and medium enterprises which cannot afford to do so on their own. As the example of the KfW suggests, it should not be taken for granted that banks cannot play a larger role in funding early-stage innovation, simply because in the U.S. that role in recent decades has been performed chiefly by specialized venture capitalists.

Finally, there is the nonprofit sector, which includes research universities.  While universities are encouraged by Bayh-Dole and other laws to back commercialization of innovations, their incentives are different from those of private venture capitalists and those incentives are subject to considerations of the larger public interest.

If the only source of capital for innovation, during the commercialization of new technologies discovered earlier by basic R&D, were venture capitalists impatient to take their profits and driven by speculative mania, then we might resign ourselves to viewing damaging financial bubbles as the regrettable but necessary price for commercializing innovation.  But speculation by private venture capitalists is only one of a number of ways to finance early-stage commercialization of breakthrough innovations.  Innovation can be funded also by existing firms, government both as procurer and state capitalist, banks and universities.

In 1822 the British essayist Charles Lamb provided a fanciful history of cooking in “A Dissertation Upon Roast Pig.”  In ancient China, he wrote, peasants discovered by accident that pigs tasted good after they were accidentally roasted when barns burned down.  For generations they cooked by burning down barns or houses.Thus, this custom of firing houses continued, till in process of time, says my manuscript, a sage arose, like our Locke, who made a discovery, that the flesh of swine, or indeed of any other animal, might be cooked (burnt, as they called it) without the necessity of consuming a whole house to dress it.[38]

The moral of this story is relevant for how we think about financial bubbles and innovation.  Just as it would be incorrect and unwise not to acknowledge that financial bubbles often accompany waves of technological innovation, it would be equally incorrect or unwise to conclude that innovation depends on such financial excess.