"The double bubble at the turn of the century: technological roots and structural implications"
This paper argues that the two boom and bust episodes of the turn of the Century -the Internet mania and crash of 1990s and the easy liquidity boom and bust of 2000s- are two distinct components of a single structural phenomenon. They are essentially the equivalent of 1929 developed in two stages, one centred on technological innovation, the other on financial innovation. Hence, the frequent references to that crash, to the 1930s and to Bretton Woods, are not simple journalistic metaphors for interpreting the "credit crunch" and its solution, but rather the intuitive recognition of a fundamental similarity between those events and the current ones. The paper holds that such major boom and bust episodes are endogenous to the way in which the market economy evolves and assimilates successive technological revolutions. It will discuss why it occurred in two bubbles on this occasion; it examines the differences and continuities between the two episodes and presents an interpretation of their nature and consequences.
1. Major Technology Bubbles as Endogenous Phenomena
The concentration on the new technologiesDecoupling and switch to quick capital gainsThe unwitting role of the MTB
2. Why the Double Bubble? The technological and historical factors
3. Two Different Bubbles: From technological to financial innovation
From opportunity pull to easy credit pushThe structural transformation in the economy4. The Underlying Continuity: The exacerbation of the casino from one boom to the nextThe bias towards financeThe double bubble and the full consequences
5. Conclusion: The special nature of major technology bubbles and the policy challenge
Table 1 Five great surges of growth and five major technology bubbles
Figure 1 Three major technology bubbles as paroxystic culmination of a long process of experimentation with new technologies and infrastructures
Figure 2 Major technology bubbles involve differential asset inflation biased to the "high tech" stocks - the information technology bubble in the 1990s
Figure 3 The mass production bubble in the 1920s was also concentrated on the high tech stocks
Figure 4 At the boom, the NASDAQ overtook the NYSE in volume of trading
Figure 5 The abandonment of fundamentals: Not earnings but capital gains
Figure 6 The abandonment of fundamentals is even more intense regarding the new technologies
Figure 7 The decoupling of the stock market from the real economy: market capitalisation disregards the behaviour of profits
Figure 8 The intensification of financial activity overtakes asset inflation during the bubble
Figure 9 Much of the increased bubble activity revolves around the new tech stocks
Figure 10 The more enduring impact of the bubble collapse on the new technology sectors than on the rest
Figure 11 The MTB also fosters the flourishing of new companies and types of funds in the financial sector
Figure 12 The late 1920s as a single major technology bubble
Figure 13 The 1990s and the 2000s: Different focus on technology or financial shares
Figure 14 The 1990s and the 2000s: Differential asset inflation
Figure 15 The 1990s and the 2000s: Change in the weight of new technology shares
Figure 16 The 1990s and the 2000s: Very different real interest rates
Figure 17 Two rhythms of monetary expansion according to Milton Friedman
Figure 18 The 1990s and the 2000s: Continuity and acceleration in the instruments of casino-type speculation
Figure 19 The intensification of globalisation after the MTB collapse and into the ELB
Figure 20 The intensification of the MTB bias towards financial profits during the ELB
Figure 21. A The decoupling from the real economy intensified from the 1990s to the 2000s
Figure 21. B The contrast with the deployment period of the previous surge: 1947-1974
The economic literature seems to pay less attention to financial bubbles than would be warranted by their profound effect on economic growth both during the boom and after the bust. There tends to be an implicit agreement that they are a derailment of the market mechanism due to external causes. In fact, the Austrian and Chicago schools, but also most neoclassical economists, tend to lay the blame on government, be it monetary policy or distorting regulation (Hayek 1933; von Mises 1949). The rational expectations school is more inclined to see such events as the intelligent work of the invisible hand, as seen in the literature on rational bubbles (Blanchard and Watson 1982; Diba and Grossman 1988).By contrast, J.K. Galbraith (1990) saw them as a recurring loop of delusion built-up by the market mechanism, but as fundamentally irrational and due to mass euphoria, herd behaviour and greed. It was Minsky (1982) -following Keynes (1936), and in turn followed by Kindleberger (1978)- who saw financial crises as a natural consequence of the way debt markets work and advanced the financial instability hypothesis.
This paper proposes to distinguish major technology bubbles (MTBs) as a special class of bubbles that constitute a recurring endogenous phenomenon, caused by the way the market economy absorbs successive technological revolutions (Perez 2002). They are different both in nature and consequences from the bubbles induced by excess liquidity from whatever source and from the Ponzi finance moments identified by Minsky. They are the result of opportunity pull rather than of easy credit push. But they are indeed bubbles. They are moments of Galbraithian irrationality, but, at least in terms of prefiguring the future value of some of the stocks involved, they also contain an element of rationality (Pastor and Veronesi 2004 and 2005).
History has given us the ideal laboratory: a major technology bubble -the 1997-2000 Internet Mania- followed by the easy liquidity bubble of 2004-07. The fact that they took place in rapid succession provides us with clearly comparable and compatible data. Yet it also suggests that they are strongly connected and interrelated.
This paper will argue that the two bubbles of the turn of the century are two stages of the same phenomenon. The first section below discusses the endogenous nature and consequences of major technology bubbles. The second analyses the reasons for the easy liquidity bubble to have followed in the wake of the NASDAQ collapse. In the third and fourth parts the two bubbles will be contrasted and compared, distinguishing their differences and similarities. Finally, there will be a brief summary of the argument and its implications in terms of policy challenges.
'...the book fills an important gap in the literature on business cycles and innovations. I most strongly commend it to all those attempting to understand the past and future evolution of technology and the economy.'
Christopher Freeman, Emeritus Professor, SPRU,
University of Sussex, UK
'...Carlota Perez shows us that historically technological revolutions arrive with remarkable regularity, and that economies react to them in predictable phases. Her argument provides much needed perspective not just on history, but on our own times. And especially on our own information revolution.'
W. Brian Arthur, Santa Fe Institute, New Mexico